nips nips2012 nips2012-81 knowledge-graph by maker-knowledge-mining

81 nips-2012-Context-Sensitive Decision Forests for Object Detection

Source: pdf

Author: Peter Kontschieder, Samuel R. Bulò, Antonio Criminisi, Pushmeet Kohli, Marcello Pelillo, Horst Bischof

Abstract: In this paper we introduce Context-Sensitive Decision Forests - A new perspective to exploit contextual information in the popular decision forest framework for the object detection problem. They are tree-structured classiﬁers with the ability to access intermediate prediction (here: classiﬁcation and regression) information during training and inference time. This intermediate prediction is available for each sample and allows us to develop context-based decision criteria, used for reﬁning the prediction process. In addition, we introduce a novel split criterion which in combination with a priority based way of constructing the trees, allows more accurate regression mode selection and hence improves the current context information. In our experiments, we demonstrate improved results for the task of pedestrian detection on the challenging TUD data set when compared to state-ofthe-art methods. 1 Introduction and Related Work In the last years, the random forest framework [1, 6] has become a very popular and powerful tool for classiﬁcation and regression problems by exhibiting many appealing properties like inherent multi-class capability, robustness to label noise and reduced tendencies to overﬁtting [7]. They are considered to be close to an ideal learner [13], making them attractive in many areas of computer vision like image classiﬁcation [5, 17], clustering [19], regression [8] or semantic segmentation [24, 15, 18]. In this work we show how the decision forest algorithm can be extended to include contextual information during learning and inference for classiﬁcation and regression problems. We focus on applying random forests to object detection, i.e. the problem of localizing multiple instances of a given object class in a test image. This task has been previously addressed in random forests [9], where the trees were modiﬁed to learn a mapping between the appearance of an image patch and its relative position to the object category centroid (i.e. center voting information). During inference, the resulting Hough Forest not only performs classiﬁcation on test samples but also casts probabilistic votes in a generalized Hough-voting space [3] that is subsequently used to obtain object center hypotheses. Ever since, a series of applications such as tracking and action recognition [10], body-joint position estimation [12] and multi-class object detection [22] have been presented. However, Hough Forests typically produce non-distinctive object hypotheses in the Hough space and hence there is the need to perform non-maximum suppression (NMS) for obtaining the ﬁnal results. While this has been addressed in [4, 26], another shortcoming is that standard (Hough) forests treat samples in a completely independent way, i.e. there is no mechanism that encourages the classiﬁer to perform consistent predictions. Within this work we are proposing that context information can be used to overcome the aforementioned problems. For example, training data for visual learning is often represented by images in form of a (regular) pixel grid topology, i.e. objects appearing in natural images can often be found in a speciﬁc context. The importance of contextual information was already highlighted in the 80’s with 1 Figure 1: Top row: Training image, label image, visualization of priority-based growing of tree (the lower, the earlier the consideration during training.). Bottom row: Inverted Hough image using [9] and breadth-ﬁrst training after 6 levels (26 = 64 nodes), Inverted Hough image after growing 64 nodes using our priority queue, Inverted Hough image using priority queue shows distinctive peaks at the end of training. a pioneering work on relaxation labelling [14] and a later work with focus on inference tasks [20] that addressed the issue of learning within the same framework. More recently, contextual information has been used in the ﬁeld of object class segmentation [21], however, mostly for high-level reasoning in random ﬁeld models or to resolve contradicting segmentation results. The introduction of contextual information as additional features in low-level classiﬁers was initially proposed in the Auto-context [25] and Semantic Texton Forest [24] models. Auto-context shows a general approach for classiﬁer boosting by iteratively learning from appearance and context information. In this line of research [18] augmented the feature space for an Entanglement Random Forest with a classiﬁcation feature, that is consequently reﬁned by the class posterior distributions according to the progress of the trained subtree. The training procedure is allowed to perform tests for speciﬁc, contextual label conﬁgurations which was demonstrated to signiﬁcantly improve the segmentation results. However, the In this paper we are presenting Context-Sensitve Decision Forests - A novel and uniﬁed interpretation of Hough Forests in light of contextual sensitivity. Our work is inspired by Auto-Context and Entanglement Forests, but instead of providing only posterior classiﬁcation results from an earlier level of the classiﬁer construction during learning and testing, we additionally provide regression (voting) information as it is used in Hough Forests. The second core contribution of our work is related to how we grow the trees: Instead of training them in a depth- or breadth-ﬁrst way, we propose a priority-based construction (which could actually consider depth- or breadth-ﬁrst as particular cases). The priority is determined by the current training error, i.e. we ﬁrst grow the parts of the tree where we experience higher error. To this end, we introduce a uniﬁed splitting criterion that estimates the joint error of classiﬁcation and regression. The consequence of using our priority-based training are illustrated in Figure 1: Given the training image with corresponding label image (top row, images 1 and 2), the tree ﬁrst tries to learn the foreground samples as shown in the color-coded plot (top row, image 3, colors correspond to index number of nodes in the tree). The effects on the intermediate prediction quality are shown in the bottom row for the regression case: The ﬁrst image shows the regression quality after training a tree with 6 levels (26 = 64 nodes) in a breadth-ﬁrst way while the second image shows the progress after growing 64 nodes according to the priority based training. Clearly, the modes for the center hypotheses are more distinctive which in turn yields to more accurate intermediate regression information that can be used for further tree construction. Our third contribution is a new family of split functions that allows to learn from training images containing multiple training instances as shown for the pedestrians in the example. We introduce a test that checks the centroid compatibility for pairs of training samples taken from the context, based on the intermediate classiﬁcation and regression derived as described before. To assess our contributions, we performed several experiments on the challenging TUD pedestrian data set [2], yielding a signiﬁcant improvement of 9% in the recall at 90% precision rate in comparison to standard Hough Forests, when learning from crowded pedestrian images. 2 2 Context-Sensitive Decision Trees This section introduces the general idea behind the context-sensitive decision forest without references to speciﬁc applications. Only in Section 3 we show a particular application to the problem of object detection. After showing some basic notational conventions that are used in the paper, we provide a section that revisits the random forest framework for classiﬁcation and regression tasks from a joint perspective, i.e. a theory allowing to consider e.g. [1, 11] and [9] in a uniﬁed way. Starting from this general view we ﬁnally introduce the context-sensitive forests in 2.2. Notations. In the paper we denote vectors using boldface lowercase (e.g. d, u, v) and sets by using uppercase calligraphic (e.g. X , Y) symbols. The sets of real, natural and integer numbers are denoted with R, N and Z as usually. We denote by 2X the power set of X and by 1 [P ] the indicator function returning 1 or 0 according to whether the proposition P is true or false. Moreover, with P(Y) we denote the set of probability distributions having Y as sample space and we implicitly assume that some σ-algebra is deﬁned on Y. We denote by δ(x) the Dirac delta function. Finally, Ex∼Q [f (x)] denotes the expectation of f (x) with respect to x sampled according to distribution Q. 2.1 Random Decision Forests for joint classiﬁcation and regression A (binary) decision tree is a tree-structured predictor1 where, starting from the root, a sample is routed until it reaches a leaf where the prediction takes place. At each internal node of the tree the decision is taken whether the sample should be forwarded to the left or right child, according to a binary-valued function. In formal terms, let X denote the input space, let Y denote the output space and let T dt be the set of decision trees. In its simplest form a decision tree consists of a single node (a leaf ) and is parametrized by a probability distribution Q ∈ P(Y) which represents the posterior probability of elements in Y given any data sample reaching the leaf. We denote this (admittedly rudimentary) tree as L F (Q) ∈ T td . Otherwise, a decision tree consists of a node with a left and a right sub-tree. This node is parametrized by a split function φ : X → {0, 1}, which determines whether to route a data sample x ∈ X reaching it to the left decision sub-tree tl ∈ T dt (if φ(x) = 0) or to the right one tr ∈ T dt (if φ(x) = 1). We denote such a tree as N D (φ, tl , tr ) ∈ T td . Finally, a decision forest is an ensemble F ⊆ T td of decision trees which makes a prediction about a data sample by averaging over the single predictions gathered from all trees. Inference. Given a decision tree t ∈ T dt , the associated posterior probability of each element in Y given a sample x ∈ X is determined by ﬁnding the probability distribution Q parametrizing the leaf that is reached by x when routed along the tree. This is compactly presented with the following deﬁnition of P (y|x, t), which is inductive in the structure of t:  if t = L F (Q) Q(y) P (y | x, t ) = P (y | x, tl ) if t = N D (φ, tl , tr ) and φ(x) = 0 (1)  P (y | x, tr ) if t = N D (φ, tl , tr ) and φ(x) = 1 . Finally, the combination of the posterior probabilities derived from the trees in a forest F ⊆ T dt can be done by an averaging operation [6], yielding a single posterior probability for the whole forest: P (y|x, F) = 1 |F| P (y|x, t) . (2) t∈F Randomized training. A random forest is created by training a set of random decision trees independently on random subsets of the training data D ⊆ X ×Y. The training procedure for a single decision tree heuristically optimizes a set of parameters like the tree structure, the split functions at the internal nodes and the density estimates at the leaves in order to reduce the prediction error on the training data. In order to prevent overﬁtting problems, the search space of possible split functions is limited to a random set and a minimum number of training samples is required to grow a leaf node. During the training procedure, each new node is fed with a set of training samples Z ⊆ D. If some stopping condition holds, depending on Z, the node becomes a leaf and a density on Y is estimated based on Z. Otherwise, an internal node is grown and a split function is selected from a pool of random ones in a way to minimize some sort of training error on Z. The selected split function induces a partition 1 we use the term predictor because we will jointly consider classiﬁcation and regression. 3 of Z into two sets, which are in turn becoming the left and right childs of the current node where the training procedure is continued, respectively. We will now write this training procedure in more formal terms. To this end we introduce a function π(Z) ∈ P(Y) providing a density on Y estimated from the training data Z ⊆ D and a loss function L(Z | Q) ∈ R penalizing wrong predictions on the training samples in Z, when predictions are given according to a distribution Q ∈ P(Y). The loss function L can be further decomposed in terms of a loss function (·|Q) : Y → R acting on each sample of the training set: L(Z | Q) = (y | Q) . (3) (x,y)∈Z Also, let Φ(Z) be a set of split functions randomly generated for a training set Z and given a split φ function φ ∈ Φ(Z), we denote by Zlφ and Zr the sets identiﬁed by splitting Z according to φ, i.e. Zlφ = {(x, y) ∈ Z : φ(x) = 0} and φ Zr = {(x, y) ∈ Z : φ(x) = 1} . We can now summarize the training procedure in terms of a recursive function g : 2X ×Y → T , which generates a random decision tree from a training set given as argument: g(Z) = L F (π(Z)) ND if some stopping condition holds φ φ, g(Zlφ ), g(Zr ) otherwise . (4) Here, we determine the optimal split function φ in the pool Φ(Z) as the one minimizing the loss we incur as a result of the node split: φ φ ∈ arg min L(Zlφ ) + L(Zr ) : φ ∈ Φ(Z) (5) where we compactly write L(Z) for L(Z|π(Z)), i.e. the loss on Z obtained with predictions driven by π(Z). A typical split function selection criterion commonly adopted for classiﬁcation and regression is information gain. The equivalent counterpart in terms of loss can be obtained by using a log-loss, i.e. (y|Q) = − log(Q(y)). A further widely used criterion is based on Gini impurity, which can be expressed in this setting by using (y|Q) = 1 − Q(y). Finally, the stopping condition that is used in (4) to determine whether to create a leaf or to continue branching the tree typically consists in checking |Z|, i.e. the number of training samples at the node, or the loss L(Z) are below some given thresholds, or if a maximum depth is reached. 2.2 Context-sensitive decision forests A context-sensitive (CS) decision tree is a decision tree in which split functions are enriched with the ability of testing contextual information of a sample, before taking a decision about where to route it. We generate contextual information at each node of a decision tree by exploiting a truncated version of the same tree as a predictor. This idea is shared with [18], however, we introduce some novelties by tackling both, classiﬁcation and regression problems in a joint manner and by leaving a wider ﬂexibility in the tree truncation procedure. We denote the set of CS decision trees as T . The main differences characterizing a CS decision tree t ∈ T compared with a standard decision tree are the following: a) every node (leaves and internal nodes) of t has an associated probability distribution Q ∈ P(Y) representing the posterior probability of an element in Y given any data sample reaching it; b) internal nodes are indexed with distinct natural numbers n ∈ N in a way to preserve the property that children nodes have a larger index compared to their parent node; c) the split function at each internal node, denoted by ϕ(·|t ) : X → {0, 1}, is bound to a CS decision tree t ∈ T , which is a truncated version of t and can be used to compute intermediate, contextual information. Similar to Section 2.1 we denote by L F (Q) ∈ T the simplest CS decision tree consisting of a single leaf node parametrized by the distribution Q, while we denote by N D (n, Q, ϕ, tl , tr ) ∈ T , the rest of the trees consisting of a node having a left and a right sub-tree, denoted by tl , tr ∈ T respectively, and being parametrized by the index n, a probability distribution Q and the split function ϕ as described above. As shown in Figure 2, the truncation of a CS decision tree at each node is obtained by exploiting the indexing imposed on the internal nodes of the tree. Given a CS decision tree t ∈ T and m ∈ N, 4 1 1 4 2 3 6 2 5 4 3 (b) The truncated version t(<5) (a) A CS decision tree t Figure 2: On the left, we ﬁnd a CS decision tree t, where only the internal nodes are indexed. On the right, we see the truncated version t(<5) of t, which is obtained by converting to leaves all nodes having index ≥ 5 (we marked with colors the corresponding node transformations). we denote by t( < τ 2 In the experiments conducted, we never exceeded 10 iterations for ﬁnding a mode. 6 (8) where Pj = P (·|(u + hj , I), t), with j = 1, 2, are the posterior probabilities obtained from tree t given samples at position u+h1 and u+h2 of image I, respectively. Please note that this test should not be confused with the regression split criterion in [9], which tries to partition the training set in a way to group examples with similar voting direction and length. Besides the novel context-sensitive split function we employ also standard split functions performing tests on X as deﬁned in [24]. 4 Experiments To assess our proposed approach, we have conducted several experiments on the task of pedestrian detection. Detecting pedestrians is very challenging for Hough-voting based methods as they typically exhibit strong articulations of feet and arms, yielding to non-distinctive hypotheses in the Hough space. We evaluated our method on the TUD pedestrian data base [2] in two different ways: First, we show our detection results with training according to the standard protocol using 400 training images (where each image contains a single annotation of a pedestrian) and evaluation on the Campus and Crossing scenes, respectively (Section 4.1). With this experiment we show the improvement over state-of-the-art approaches when learning can be performed with simultaneous knowledge about context information. In a second variation (Section 4.2), we use the images of the Crossing scene (201 images) as a training set. Most images of this scene contain more than four persons with strong overlap and mutual occlusions. However, instead of using the original annotation which covers only pedestrians with at least 50% overlap (1008 bounding boxes), we use the more accurate, pixel-wise ground truth annotations of [23] for the entire scene that includes all persons and consists of 1215 bounding boxes. Please note that this annotation is even more detailed than the one presented in [4] with 1018 bounding boxes. The purpose of the second experiment is to show that our context-sensitive forest can exploit the availability of multiple training instances signiﬁcantly better than state-of-the-art. The most related work and therefore also the baseline in our experiments is the Hough Forest [9]. To guarantee a fair comparison, we use the same training parameters for [9] and our context sensitive forest: We trained 20 trees and the training data (including horizontally ﬂipped images) was sampled homogeneously per category per image. The patch size was ﬁxed to 30 × 30 and we performed 1600 node tests for ﬁnding the best split function parameters per node. The trees were stopped growing when < 7 samples were available. As image features, we used the the ﬁrst 16 feature channels provided in the publicly available Hough Forest code of [9]. In order to obtain the object detection hypotheses from the Hough space, we use the same Non-maximum suppression (NMS) technique in all our experiments as suggested in [9]. To evaluate the obtained hypotheses, we use the standard PASAL-VOC criterion which requires the mutual overlap between ground truth and detected bounding boxes to be ≥ 50%. The additional parameter of (7) was ﬁxed to σ = 7. 4.1 Evaluation using standard protocol training set The standard training set contains 400 images where each image comes with a single pedestrian annotation. For our experiments, we rescaled the images by a factor of 0.5 and doubled the training image set by including also the horizontally ﬂipped images. We randomly chose 125 training samples per image for foreground and background, resulting in 2 · 400 · 2 · 125 = 200k training samples per tree. For additional comparisons, we provide the results presented in the recent work on joint object detection and segmentation of [23], from which we also provide evaluation results of the Implicit Shape Model (ISM) [16]. However, please note that the results of [23] are based on a different baseline implementation. Moreover, we show the results of [4] when using the provided code and conﬁguration ﬁles from the ﬁrst authors homepage. Unfortunately, we could not reproduce the results of the original paper. First, we discuss the results obtained on the Campus scene. This data set consists of 71 images showing walking pedestrians at severe scale differences and partial occlusions. The ground truth we use has been released with [4] and contains a total number of 314 pedestrians. Figure 3, ﬁrst row, plot 1 shows the precision-recall curves when using 3 scales (factors 0.3, 0.4, 0.55) for our baseline [9] (blue), results from re-evaluating [4] (cyan, 5 scales), [23] (green) and our ContextSensitive Forest without and with using the priority queue based tree construction (red/magenta). In case of not using the priority queue, we trained the trees according to a breadth-ﬁrst way. We obtain a performance boost of ≈ 6% in recall at a precision of 90% when using both, context information and the priority based construction of our forest. The second plot in the ﬁrst row of Figure 3 shows the results when the same forests are tested on the Crossing scene, using the more detailed ground 7 TUD Campus (3 scales) TUD−Crossing (3 scales) 0.9 0.8 0.8 0.7 0.7 0.6 0.6 Precision 1 0.9 Precision 1 0.5 0.4 0.3 0.2 0.1 0 0 0.5 0.4 0.3 Baseline Hough Forest Barinova et al. CVPR’10, 5 scales Proposed Context−Sensitive, No Priority Queue Proposed Context−Sensitive, With Priority Queue Riemenschneider et al. ECCV’12 0.1 0.2 0.3 0.4 0.5 Recall 0.6 0.7 0.8 0.2 0.1 0.9 0 0 1 Baseline Hough Forest Barinova et al. CVPR’10 Proposed Context−Sensitive, No Priority Queue Proposed Context−Sensitive, With Priority Queue Riemenschneider et al. ECCV’12 (1 scale) Leibe et al. IJCV’08 (1 scale) 0.1 TUD Campus (3 scales) 0.3 0.4 0.5 Recall 0.6 0.7 0.8 0.9 1 0.9 1 1 0.9 0.8 0.8 0.7 0.7 0.6 0.6 Precision 1 0.9 Precision 0.2 TUD Campus (5 scales) 0.5 0.4 0.3 0 0 0.4 0.3 0.2 0.1 0.5 0.2 Baseline Hough Forest Proposed Context−Sensitive, No Priority Queue Proposed Context−Sensitive, With Priority Queue 0.1 0.2 0.3 0.4 0.5 Recall 0.6 0.7 0.8 0.1 0.9 1 0 0 Baseline Hough Forest Proposed Context−Sensitive, No Priority Queue Proposed Context−Sensitive, With Priority Queue 0.1 0.2 0.3 0.4 0.5 Recall 0.6 0.7 0.8 Figure 3: Precision-Recall Curves for detections, Top row: Standard training (400 images), evaluation on Campus and Crossing (3 scales). Bottom row: Training on Crossing annotations of [23], evaluation on Campus, 3 and 5 scales. Right images: Qualitative examples for Campus (top 2) and Crossing (bottom 2) scenes. (green) correctly found by our method (blue) ground truth (red) wrong association (cyan) missed detection. truth annotations. The data set shows walking pedestrians (Figure 3, right side, last 2 images) with a smaller variation in scale compared to the Campus scene but with strong mutual occlusions and overlaps. The improvement with respect to the baseline is lower (≈ 2% gain at a precision of 90%) and we ﬁnd similar developments of the curves. However, this comes somewhat expectedly as the training data does not properly reﬂect the occlusions we actually want to model. 4.2 Evaluation on Campus scene using Crossing scene as training set In our next experiment we trained the forests (same parameters) on the novel annotations of [23] for the Crossing scene. Please note that this reduces the training set to only 201 images (we did not include the ﬂipped images). Qualitative detection results are shown in Figure 3, right side, images 1 and 2. From the ﬁrst precison-recall curve in the second row of Figure 3 we can see, that the margin between the baseline and our proposed method could be clearly improved (gain of ≈ 9% recall at precision 90%) when evaluating on the same 3 scales. With evaluation on 5 scales (factors 0.34, 0.42, 0.51, 0.65, 0.76) we found a strong increase in the recall, however, at the cost of loosing 2 − 3% of precision below a recall of 60%, as illustrated in the second plot of row 2 in Figure 3. While our method is able to maintain a precision above 90% up to a recall of ≈ 83%, the baseline implementation drops already at a recall of ≈ 20%. 5 Conclusions In this work we have presented Context-Sensitive Decision Forests with application to the object detection problem. Our new forest has the ability to access intermediate prediction (classiﬁcation and regression) information about all samples of the training set and can therefore learn from contextual information throughout the growing process. This is in contrast to existing random forest methods used for object detection which typically treat training samples in an independent manner. Moreover, we have introduced a novel splitting criterion together with a mode isolation technique, which allows us to (a) perform a priority-driven way of tree growing and (b) install novel context-based test functions to check for mutual object centroid agreements. In our experimental results on pedestrian detection we demonstrated superior performance with respect to state-of-the-art methods and additionally found that our new algorithm can signiﬁcantly better exploit training data containing multiple training objects. Acknowledgements. Peter Kontschieder acknowledges ﬁnancial support of the Austrian Science Fund (FWF) from project ’Fibermorph’ with number P22261-N22. 8 References [1] Y. Amit and D. Geman. Shape quantization and recognition with randomized trees. Neural Computation, 1997. [2] M. Andriluka, S. Roth, and B. Schiele. People-tracking-by-detection and people-detection-by-tracking. In (CVPR), 2008. [3] D. H. Ballard. Generalizing the hough transform to detect arbitrary shapes. Pattern Recognition, 13(2), 1981. [4] O. Barinova, V. Lempitsky, and P. Kohli. On detection of multiple object instances using hough transforms. In (CVPR), 2010. [5] A. Bosch, A. Zisserman, and X. Mu˜oz. Image classiﬁcation using random forests and ferns. In (ICCV), n 2007. [6] L. Breiman. Random forests. In Machine Learning, 2001. [7] A. Criminisi, J. Shotton, and E. Konukoglu. Decision forests: A uniﬁed framework for classiﬁcation, regression, density estimation, manifold learning and semi-supervised learning. In Foundations and Trends in Computer Graphics and Vision, volume 7, pages 81–227, 2012. [8] A. Criminisi, J. Shotton, D. Robertson, and E. Konukoglu. Regression forests for efﬁcient anatomy detection and localization in CT scans. In MICCAI-MCV Workshop, 2010. [9] J. Gall and V. Lempitsky. Class-speciﬁc hough forests for object detection. In (CVPR), 2009. [10] J. Gall, A. Yao, N. Razavi, L. Van Gool, and V. Lempitsky. Hough forests for object detection, tracking, and action recognition. (PAMI), 2011. [11] P. Geurts, D. Ernst, and L. Wehenkel. Extremely randomized trees. Machine Learning, 2006. [12] R. Girshick, J. Shotton, P. Kohli, A. Criminisi, and A. Fitzgibbon. Efﬁcient regression of general-activity human poses from depth images. In (ICCV), 2011. [13] T. Hastie, R. Tibshirani, and J. H. Friedman. The Elements of Statistical Learning. Springer, 2009. [14] R. A. Hummel and S. W. Zucker. On the foundations of relaxation labeling. (PAMI), 5(3):267–287, 1983. [15] P. Kontschieder, S. Rota Bul` , H. Bischof, and M. Pelillo. Structured class-labels in random forests for o semantic image labelling. In (ICCV), 2011. [16] B. Leibe, A. Leonardis, and B. Schiele. Robust object detection with interleaved categorization and segmentation. (IJCV), 2008. [17] R. Mar´ e, P. Geurts, J. Piater, and L. Wehenkel. Random subwindows for robust image classiﬁcation. In e (CVPR), 2005. [18] A. Montillo, J. Shotton, J. Winn, J. E. Iglesias, D. Metaxas, and A. Criminisi. Entangled decision forests and their application for semantic segmentation of CT images. In (IPMI), 2011. [19] F. Moosmann, B. Triggs, and F. Jurie. Fast discriminative visual codebooks using randomized clustering forests. In (NIPS), 2006. [20] M. Pelillo and M. Reﬁce. Learning compatibility coefﬁcients for relaxation labeling processes. (PAMI), 16(9):933–945, 1994. [21] A. Rabinovich, A. Vedaldi, C. Galleguillos, E. Wiewiora, and S. Belongie. Objects in context. In (ICCV), 2007. [22] N. Razavi, J. Gall, and L. Van Gool. Scalable multi-class object detection. In (CVPR), 2011. [23] H. Riemenschneider, S. Sternig, M. Donoser, P. M. Roth, and H. Bischof. Hough regions for joining instance localization and segmentation. In (ECCV), 2012. [24] J. Shotton, M. Johnson, and R. Cipolla. Semantic texton forests for image categorization and segmentation. In (CVPR), 2008. [25] Z. Tu. Auto-context and its application to high-level vision tasks. In (CVPR), 2008. [26] O. Woodford, M. Pham, A. Maki, F. Perbet, and B. Stenger. Demisting the hough transform for 3d shape recognition and registration. In (BMVC), 2011. 9

Reference: text

Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 They are tree-structured classiﬁers with the ability to access intermediate prediction (here: classiﬁcation and regression) information during training and inference time. [sent-2, score-0.194]

2 This intermediate prediction is available for each sample and allows us to develop context-based decision criteria, used for reﬁning the prediction process. [sent-3, score-0.316]

3 In addition, we introduce a novel split criterion which in combination with a priority based way of constructing the trees, allows more accurate regression mode selection and hence improves the current context information. [sent-4, score-0.544]

4 In our experiments, we demonstrate improved results for the task of pedestrian detection on the challenging TUD data set when compared to state-ofthe-art methods. [sent-5, score-0.237]

5 They are considered to be close to an ideal learner [13], making them attractive in many areas of computer vision like image classiﬁcation [5, 17], clustering [19], regression [8] or semantic segmentation [24, 15, 18]. [sent-7, score-0.257]

6 In this work we show how the decision forest algorithm can be extended to include contextual information during learning and inference for classiﬁcation and regression problems. [sent-8, score-0.683]

7 We focus on applying random forests to object detection, i. [sent-9, score-0.434]

8 This task has been previously addressed in random forests [9], where the trees were modiﬁed to learn a mapping between the appearance of an image patch and its relative position to the object category centroid (i. [sent-12, score-0.656]

9 Ever since, a series of applications such as tracking and action recognition [10], body-joint position estimation [12] and multi-class object detection [22] have been presented. [sent-16, score-0.208]

10 However, Hough Forests typically produce non-distinctive object hypotheses in the Hough space and hence there is the need to perform non-maximum suppression (NMS) for obtaining the ﬁnal results. [sent-17, score-0.188]

11 While this has been addressed in [4, 26], another shortcoming is that standard (Hough) forests treat samples in a completely independent way, i. [sent-18, score-0.367]

12 For example, training data for visual learning is often represented by images in form of a (regular) pixel grid topology, i. [sent-22, score-0.196]

13 The importance of contextual information was already highlighted in the 80’s with 1 Figure 1: Top row: Training image, label image, visualization of priority-based growing of tree (the lower, the earlier the consideration during training. [sent-25, score-0.402]

14 Bottom row: Inverted Hough image using [9] and breadth-ﬁrst training after 6 levels (26 = 64 nodes), Inverted Hough image after growing 64 nodes using our priority queue, Inverted Hough image using priority queue shows distinctive peaks at the end of training. [sent-27, score-1.166]

15 More recently, contextual information has been used in the ﬁeld of object class segmentation [21], however, mostly for high-level reasoning in random ﬁeld models or to resolve contradicting segmentation results. [sent-29, score-0.355]

16 The introduction of contextual information as additional features in low-level classiﬁers was initially proposed in the Auto-context [25] and Semantic Texton Forest [24] models. [sent-30, score-0.136]

17 The training procedure is allowed to perform tests for speciﬁc, contextual label conﬁgurations which was demonstrated to signiﬁcantly improve the segmentation results. [sent-33, score-0.302]

18 However, the In this paper we are presenting Context-Sensitve Decision Forests - A novel and uniﬁed interpretation of Hough Forests in light of contextual sensitivity. [sent-34, score-0.136]

19 The second core contribution of our work is related to how we grow the trees: Instead of training them in a depth- or breadth-ﬁrst way, we propose a priority-based construction (which could actually consider depth- or breadth-ﬁrst as particular cases). [sent-36, score-0.136]

20 The priority is determined by the current training error, i. [sent-37, score-0.349]

21 we ﬁrst grow the parts of the tree where we experience higher error. [sent-39, score-0.236]

22 Clearly, the modes for the center hypotheses are more distinctive which in turn yields to more accurate intermediate regression information that can be used for further tree construction. [sent-43, score-0.389]

23 Our third contribution is a new family of split functions that allows to learn from training images containing multiple training instances as shown for the pedestrians in the example. [sent-44, score-0.516]

24 We introduce a test that checks the centroid compatibility for pairs of training samples taken from the context, based on the intermediate classiﬁcation and regression derived as described before. [sent-45, score-0.348]

25 To assess our contributions, we performed several experiments on the challenging TUD pedestrian data set [2], yielding a signiﬁcant improvement of 9% in the recall at 90% precision rate in comparison to standard Hough Forests, when learning from crowded pedestrian images. [sent-46, score-0.366]

26 2 2 Context-Sensitive Decision Trees This section introduces the general idea behind the context-sensitive decision forest without references to speciﬁc applications. [sent-47, score-0.475]

27 After showing some basic notational conventions that are used in the paper, we provide a section that revisits the random forest framework for classiﬁcation and regression tasks from a joint perspective, i. [sent-49, score-0.346]

28 Starting from this general view we ﬁnally introduce the context-sensitive forests in 2. [sent-54, score-0.333]

29 1 Random Decision Forests for joint classiﬁcation and regression A (binary) decision tree is a tree-structured predictor1 where, starting from the root, a sample is routed until it reaches a leaf where the prediction takes place. [sent-68, score-0.635]

30 At each internal node of the tree the decision is taken whether the sample should be forwarded to the left or right child, according to a binary-valued function. [sent-69, score-0.586]

31 In formal terms, let X denote the input space, let Y denote the output space and let T dt be the set of decision trees. [sent-70, score-0.243]

32 In its simplest form a decision tree consists of a single node (a leaf ) and is parametrized by a probability distribution Q ∈ P(Y) which represents the posterior probability of elements in Y given any data sample reaching the leaf. [sent-71, score-0.713]

33 We denote this (admittedly rudimentary) tree as L F (Q) ∈ T td . [sent-72, score-0.243]

34 Otherwise, a decision tree consists of a node with a left and a right sub-tree. [sent-73, score-0.52]

35 This node is parametrized by a split function φ : X → {0, 1}, which determines whether to route a data sample x ∈ X reaching it to the left decision sub-tree tl ∈ T dt (if φ(x) = 0) or to the right one tr ∈ T dt (if φ(x) = 1). [sent-74, score-0.788]

36 We denote such a tree as N D (φ, tl , tr ) ∈ T td . [sent-75, score-0.406]

37 Finally, a decision forest is an ensemble F ⊆ T td of decision trees which makes a prediction about a data sample by averaging over the single predictions gathered from all trees. [sent-76, score-0.83]

38 Given a decision tree t ∈ T dt , the associated posterior probability of each element in Y given a sample x ∈ X is determined by ﬁnding the probability distribution Q parametrizing the leaf that is reached by x when routed along the tree. [sent-78, score-0.614]

39 This is compactly presented with the following deﬁnition of P (y|x, t), which is inductive in the structure of t:  if t = L F (Q) Q(y) P (y | x, t ) = P (y | x, tl ) if t = N D (φ, tl , tr ) and φ(x) = 0 (1)  P (y | x, tr ) if t = N D (φ, tl , tr ) and φ(x) = 1 . [sent-79, score-0.489]

40 Finally, the combination of the posterior probabilities derived from the trees in a forest F ⊆ T dt can be done by an averaging operation [6], yielding a single posterior probability for the whole forest: P (y|x, F) = 1 |F| P (y|x, t) . [sent-80, score-0.48]

41 A random forest is created by training a set of random decision trees independently on random subsets of the training data D ⊆ X ×Y. [sent-82, score-0.779]

42 The training procedure for a single decision tree heuristically optimizes a set of parameters like the tree structure, the split functions at the internal nodes and the density estimates at the leaves in order to reduce the prediction error on the training data. [sent-83, score-1.113]

43 In order to prevent overﬁtting problems, the search space of possible split functions is limited to a random set and a minimum number of training samples is required to grow a leaf node. [sent-84, score-0.374]

44 During the training procedure, each new node is fed with a set of training samples Z ⊆ D. [sent-85, score-0.36]

45 If some stopping condition holds, depending on Z, the node becomes a leaf and a density on Y is estimated based on Z. [sent-86, score-0.224]

46 Otherwise, an internal node is grown and a split function is selected from a pool of random ones in a way to minimize some sort of training error on Z. [sent-87, score-0.409]

47 3 of Z into two sets, which are in turn becoming the left and right childs of the current node where the training procedure is continued, respectively. [sent-89, score-0.219]

48 To this end we introduce a function π(Z) ∈ P(Y) providing a density on Y estimated from the training data Z ⊆ D and a loss function L(Z | Q) ∈ R penalizing wrong predictions on the training samples in Z, when predictions are given according to a distribution Q ∈ P(Y). [sent-91, score-0.248]

49 (3) (x,y)∈Z Also, let Φ(Z) be a set of split functions randomly generated for a training set Z and given a split φ function φ ∈ Φ(Z), we denote by Zlφ and Zr the sets identiﬁed by splitting Z according to φ, i. [sent-93, score-0.385]

50 We can now summarize the training procedure in terms of a recursive function g : 2X ×Y → T , which generates a random decision tree from a training set given as argument: g(Z) = L F (π(Z)) ND if some stopping condition holds φ φ, g(Zlφ ), g(Zr ) otherwise . [sent-96, score-0.654]

51 (4) Here, we determine the optimal split function φ in the pool Φ(Z) as the one minimizing the loss we incur as a result of the node split: φ φ ∈ arg min L(Zlφ ) + L(Zr ) : φ ∈ Φ(Z) (5) where we compactly write L(Z) for L(Z|π(Z)), i. [sent-97, score-0.236]

52 A typical split function selection criterion commonly adopted for classiﬁcation and regression is information gain. [sent-100, score-0.242]

53 Finally, the stopping condition that is used in (4) to determine whether to create a leaf or to continue branching the tree typically consists in checking |Z|, i. [sent-105, score-0.319]

54 the number of training samples at the node, or the loss L(Z) are below some given thresholds, or if a maximum depth is reached. [sent-107, score-0.141]

55 2 Context-sensitive decision forests A context-sensitive (CS) decision tree is a decision tree in which split functions are enriched with the ability of testing contextual information of a sample, before taking a decision about where to route it. [sent-109, score-1.839]

56 We generate contextual information at each node of a decision tree by exploiting a truncated version of the same tree as a predictor. [sent-110, score-0.901]

57 This idea is shared with [18], however, we introduce some novelties by tackling both, classiﬁcation and regression problems in a joint manner and by leaving a wider ﬂexibility in the tree truncation procedure. [sent-111, score-0.279]

58 As shown in Figure 2, the truncation of a CS decision tree at each node is obtained by exploiting the indexing imposed on the internal nodes of the tree. [sent-116, score-0.652]

59 Given a CS decision tree t ∈ T and m ∈ N, 4 1 1 4 2 3 6 2 5 4 3 (b) The truncated version t(<5) (a) A CS decision tree t Figure 2: On the left, we ﬁnd a CS decision tree t, where only the internal nodes are indexed. [sent-117, score-1.394]

60 On the right, we see the truncated version t(<5) of t, which is obtained by converting to leaves all nodes having index ≥ 5 (we marked with colors the corresponding node transformations). [sent-118, score-0.244]

61 6 (8) where Pj = P (·|(u + hj , I), t), with j = 1, 2, are the posterior probabilities obtained from tree t given samples at position u+h1 and u+h2 of image I, respectively. [sent-120, score-0.363]

62 Please note that this test should not be confused with the regression split criterion in [9], which tries to partition the training set in a way to group examples with similar voting direction and length. [sent-121, score-0.384]

63 Besides the novel context-sensitive split function we employ also standard split functions performing tests on X as deﬁned in [24]. [sent-122, score-0.248]

64 Detecting pedestrians is very challenging for Hough-voting based methods as they typically exhibit strong articulations of feet and arms, yielding to non-distinctive hypotheses in the Hough space. [sent-124, score-0.14]

65 2), we use the images of the Crossing scene (201 images) as a training set. [sent-129, score-0.256]

66 Most images of this scene contain more than four persons with strong overlap and mutual occlusions. [sent-130, score-0.241]

67 However, instead of using the original annotation which covers only pedestrians with at least 50% overlap (1008 bounding boxes), we use the more accurate, pixel-wise ground truth annotations of [23] for the entire scene that includes all persons and consists of 1215 bounding boxes. [sent-131, score-0.421]

68 The purpose of the second experiment is to show that our context-sensitive forest can exploit the availability of multiple training instances signiﬁcantly better than state-of-the-art. [sent-133, score-0.381]

69 To guarantee a fair comparison, we use the same training parameters for [9] and our context sensitive forest: We trained 20 trees and the training data (including horizontally ﬂipped images) was sampled homogeneously per category per image. [sent-135, score-0.467]

70 The patch size was ﬁxed to 30 × 30 and we performed 1600 node tests for ﬁnding the best split function parameters per node. [sent-136, score-0.236]

71 The trees were stopped growing when < 7 samples were available. [sent-137, score-0.183]

72 In order to obtain the object detection hypotheses from the Hough space, we use the same Non-maximum suppression (NMS) technique in all our experiments as suggested in [9]. [sent-139, score-0.295]

73 To evaluate the obtained hypotheses, we use the standard PASAL-VOC criterion which requires the mutual overlap between ground truth and detected bounding boxes to be ≥ 50%. [sent-140, score-0.214]

74 1 Evaluation using standard protocol training set The standard training set contains 400 images where each image comes with a single pedestrian annotation. [sent-143, score-0.518]

75 5 and doubled the training image set by including also the horizontally ﬂipped images. [sent-145, score-0.228]

76 We randomly chose 125 training samples per image for foreground and background, resulting in 2 · 400 · 2 · 125 = 200k training samples per tree. [sent-146, score-0.401]

77 For additional comparisons, we provide the results presented in the recent work on joint object detection and segmentation of [23], from which we also provide evaluation results of the Implicit Shape Model (ISM) [16]. [sent-147, score-0.267]

78 This data set consists of 71 images showing walking pedestrians at severe scale differences and partial occlusions. [sent-152, score-0.209]

79 55) for our baseline [9] (blue), results from re-evaluating [4] (cyan, 5 scales), [23] (green) and our ContextSensitive Forest without and with using the priority queue based tree construction (red/magenta). [sent-157, score-0.706]

80 In case of not using the priority queue, we trained the trees according to a breadth-ﬁrst way. [sent-158, score-0.332]

81 We obtain a performance boost of ≈ 6% in recall at a precision of 90% when using both, context information and the priority based construction of our forest. [sent-159, score-0.408]

82 The second plot in the ﬁrst row of Figure 3 shows the results when the same forests are tested on the Crossing scene, using the more detailed ground 7 TUD Campus (3 scales) TUD−Crossing (3 scales) 0. [sent-160, score-0.417]

83 The data set shows walking pedestrians (Figure 3, right side, last 2 images) with a smaller variation in scale compared to the Campus scene but with strong mutual occlusions and overlaps. [sent-242, score-0.242]

84 2 Evaluation on Campus scene using Crossing scene as training set In our next experiment we trained the forests (same parameters) on the novel annotations of [23] for the Crossing scene. [sent-246, score-0.6]

85 Please note that this reduces the training set to only 201 images (we did not include the ﬂipped images). [sent-247, score-0.196]

86 Qualitative detection results are shown in Figure 3, right side, images 1 and 2. [sent-248, score-0.196]

87 From the ﬁrst precison-recall curve in the second row of Figure 3 we can see, that the margin between the baseline and our proposed method could be clearly improved (gain of ≈ 9% recall at precision 90%) when evaluating on the same 3 scales. [sent-249, score-0.216]

88 76) we found a strong increase in the recall, however, at the cost of loosing 2 − 3% of precision below a recall of 60%, as illustrated in the second plot of row 2 in Figure 3. [sent-255, score-0.154]

89 While our method is able to maintain a precision above 90% up to a recall of ≈ 83%, the baseline implementation drops already at a recall of ≈ 20%. [sent-256, score-0.213]

90 5 Conclusions In this work we have presented Context-Sensitive Decision Forests with application to the object detection problem. [sent-257, score-0.208]

91 Our new forest has the ability to access intermediate prediction (classiﬁcation and regression) information about all samples of the training set and can therefore learn from contextual information throughout the growing process. [sent-258, score-0.697]

92 This is in contrast to existing random forest methods used for object detection which typically treat training samples in an independent manner. [sent-259, score-0.623]

93 Moreover, we have introduced a novel splitting criterion together with a mode isolation technique, which allows us to (a) perform a priority-driven way of tree growing and (b) install novel context-based test functions to check for mutual object centroid agreements. [sent-260, score-0.523]

94 In our experimental results on pedestrian detection we demonstrated superior performance with respect to state-of-the-art methods and additionally found that our new algorithm can signiﬁcantly better exploit training data containing multiple training objects. [sent-261, score-0.451]

95 On detection of multiple object instances using hough transforms. [sent-284, score-0.598]

96 Regression forests for efﬁcient anatomy detection and localization in CT scans. [sent-307, score-0.44]

97 Hough forests for object detection, tracking, and action recognition. [sent-320, score-0.434]

98 Structured class-labels in random forests for o semantic image labelling. [sent-355, score-0.459]

99 Entangled decision forests and their application for semantic segmentation of CT images. [sent-378, score-0.634]

100 Demisting the hough transform for 3d shape recognition and registration. [sent-430, score-0.39]

similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('hough', 0.39), ('forests', 0.333), ('forest', 0.274), ('priority', 0.242), ('tree', 0.207), ('decision', 0.201), ('queue', 0.195), ('campus', 0.17), ('tud', 0.163), ('crossing', 0.141), ('contextual', 0.136), ('pedestrian', 0.13), ('split', 0.124), ('tl', 0.114), ('node', 0.112), ('detection', 0.107), ('training', 0.107), ('object', 0.101), ('cs', 0.09), ('trees', 0.09), ('pedestrians', 0.089), ('shotton', 0.089), ('images', 0.089), ('image', 0.085), ('zr', 0.082), ('leaf', 0.08), ('regression', 0.072), ('riemenschneider', 0.07), ('zl', 0.068), ('sensitive', 0.067), ('nodes', 0.066), ('internal', 0.066), ('baseline', 0.062), ('barinova', 0.062), ('criminisi', 0.062), ('gall', 0.062), ('precision', 0.061), ('cvpr', 0.06), ('context', 0.06), ('scene', 0.06), ('intermediate', 0.059), ('segmentation', 0.059), ('growing', 0.059), ('ipped', 0.054), ('classi', 0.053), ('hypotheses', 0.051), ('tr', 0.049), ('inverted', 0.049), ('row', 0.048), ('please', 0.048), ('scales', 0.048), ('centroid', 0.047), ('bul', 0.047), ('entanglement', 0.047), ('kontschieder', 0.047), ('nms', 0.047), ('razavi', 0.047), ('rota', 0.047), ('routed', 0.047), ('criterion', 0.046), ('recall', 0.045), ('parametrized', 0.043), ('dt', 0.042), ('semantic', 0.041), ('texton', 0.041), ('annotations', 0.04), ('truncated', 0.038), ('truth', 0.038), ('posterior', 0.037), ('td', 0.036), ('ground', 0.036), ('horizontally', 0.036), ('geurts', 0.036), ('suppression', 0.036), ('randomized', 0.035), ('voting', 0.035), ('samples', 0.034), ('foreground', 0.034), ('iccv', 0.033), ('mutual', 0.033), ('annotation', 0.033), ('reaching', 0.033), ('bounding', 0.033), ('leibe', 0.033), ('uni', 0.032), ('stopping', 0.032), ('persons', 0.031), ('cyan', 0.031), ('walking', 0.031), ('pami', 0.031), ('splitting', 0.03), ('compatibility', 0.029), ('occlusions', 0.029), ('eccv', 0.029), ('grow', 0.029), ('prediction', 0.028), ('overlap', 0.028), ('route', 0.028), ('colors', 0.028), ('ijcv', 0.027)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 1.0000006 81 nips-2012-Context-Sensitive Decision Forests for Object Detection

Author: Peter Kontschieder, Samuel R. Bulò, Antonio Criminisi, Pushmeet Kohli, Marcello Pelillo, Horst Bischof

2 0.16161849 232 nips-2012-Multiplicative Forests for Continuous-Time Processes

Author: Jeremy Weiss, Sriraam Natarajan, David Page

Abstract: Learning temporal dependencies between variables over continuous time is an important and challenging task. Continuous-time Bayesian networks effectively model such processes but are limited by the number of conditional intensity matrices, which grows exponentially in the number of parents per variable. We develop a partition-based representation using regression trees and forests whose parameter spaces grow linearly in the number of node splits. Using a multiplicative assumption we show how to update the forest likelihood in closed form, producing efﬁcient model updates. Our results show multiplicative forests can be learned from few temporal trajectories with large gains in performance and scalability.

3 0.13525754 306 nips-2012-Semantic Kernel Forests from Multiple Taxonomies

Author: Sung J. Hwang, Kristen Grauman, Fei Sha

Abstract: When learning features for complex visual recognition problems, labeled image exemplars alone can be insufﬁcient. While an object taxonomy specifying the categories’ semantic relationships could bolster the learning process, not all relationships are relevant to a given visual classiﬁcation task, nor does a single taxonomy capture all ties that are relevant. In light of these issues, we propose a discriminative feature learning approach that leverages multiple hierarchical taxonomies representing different semantic views of the object categories (e.g., for animal classes, one taxonomy could reﬂect their phylogenic ties, while another could reﬂect their habitats). For each taxonomy, we ﬁrst learn a tree of semantic kernels, where each node has a Mahalanobis kernel optimized to distinguish between the classes in its children nodes. Then, using the resulting semantic kernel forest, we learn class-speciﬁc kernel combinations to select only those relationships relevant to recognize each object class. To learn the weights, we introduce a novel hierarchical regularization term that further exploits the taxonomies’ structure. We demonstrate our method on challenging object recognition datasets, and show that interleaving multiple taxonomic views yields signiﬁcant accuracy improvements.

4 0.1337578 344 nips-2012-Timely Object Recognition

Author: Sergey Karayev, Tobias Baumgartner, Mario Fritz, Trevor Darrell

Abstract: In a large visual multi-class detection framework, the timeliness of results can be crucial. Our method for timely multi-class detection aims to give the best possible performance at any single point after a start time; it is terminated at a deadline time. Toward this goal, we formulate a dynamic, closed-loop policy that infers the contents of the image in order to decide which detector to deploy next. In contrast to previous work, our method signiﬁcantly diverges from the predominant greedy strategies, and is able to learn to take actions with deferred values. We evaluate our method with a novel timeliness measure, computed as the area under an Average Precision vs. Time curve. Experiments are conducted on the PASCAL VOC object detection dataset. If execution is stopped when only half the detectors have been run, our method obtains 66% better AP than a random ordering, and 14% better performance than an intelligent baseline. On the timeliness measure, our method obtains at least 11% better performance. Our method is easily extensible, as it treats detectors and classiﬁers as black boxes and learns from execution traces using reinforcement learning. 1

5 0.12440757 339 nips-2012-The Time-Marginalized Coalescent Prior for Hierarchical Clustering

Author: Levi Boyles, Max Welling

Abstract: We introduce a new prior for use in Nonparametric Bayesian Hierarchical Clustering. The prior is constructed by marginalizing out the time information of Kingman’s coalescent, providing a prior over tree structures which we call the Time-Marginalized Coalescent (TMC). This allows for models which factorize the tree structure and times, providing two beneﬁts: more ﬂexible priors may be constructed and more efﬁcient Gibbs type inference can be used. We demonstrate this on an example model for density estimation and show the TMC achieves competitive experimental results. 1

6 0.10734544 260 nips-2012-Online Sum-Product Computation Over Trees

7 0.10629623 106 nips-2012-Dynamical And-Or Graph Learning for Object Shape Modeling and Detection

8 0.10586919 172 nips-2012-Latent Graphical Model Selection: Efficient Methods for Locally Tree-like Graphs

9 0.1020058 209 nips-2012-Max-Margin Structured Output Regression for Spatio-Temporal Action Localization

10 0.096157312 200 nips-2012-Local Supervised Learning through Space Partitioning

11 0.095199883 105 nips-2012-Dynamic Pruning of Factor Graphs for Maximum Marginal Prediction

12 0.093508504 8 nips-2012-A Generative Model for Parts-based Object Segmentation

13 0.089935236 40 nips-2012-Analyzing 3D Objects in Cluttered Images

14 0.089858264 180 nips-2012-Learning Mixtures of Tree Graphical Models

15 0.088962846 183 nips-2012-Learning Partially Observable Models Using Temporally Abstract Decision Trees

16 0.08697705 97 nips-2012-Diffusion Decision Making for Adaptive k-Nearest Neighbor Classification

17 0.08557266 303 nips-2012-Searching for objects driven by context

18 0.082495652 201 nips-2012-Localizing 3D cuboids in single-view images

19 0.077039689 178 nips-2012-Learning Label Trees for Probabilistic Modelling of Implicit Feedback

20 0.076537609 92 nips-2012-Deep Representations and Codes for Image Auto-Annotation

similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.201), (1, 0.004), (2, -0.142), (3, -0.048), (4, 0.047), (5, -0.125), (6, -0.002), (7, -0.052), (8, -0.163), (9, 0.039), (10, -0.072), (11, 0.051), (12, 0.139), (13, -0.101), (14, -0.058), (15, 0.051), (16, -0.04), (17, 0.003), (18, 0.012), (19, -0.046), (20, -0.038), (21, 0.135), (22, 0.064), (23, 0.049), (24, -0.027), (25, -0.031), (26, -0.021), (27, -0.027), (28, 0.07), (29, 0.021), (30, -0.004), (31, -0.101), (32, -0.053), (33, -0.12), (34, -0.049), (35, 0.143), (36, -0.06), (37, 0.044), (38, -0.038), (39, -0.003), (40, 0.024), (41, -0.055), (42, 0.039), (43, -0.034), (44, 0.049), (45, -0.022), (46, -0.072), (47, 0.019), (48, -0.005), (49, -0.006)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.95234329 81 nips-2012-Context-Sensitive Decision Forests for Object Detection

Author: Peter Kontschieder, Samuel R. Bulò, Antonio Criminisi, Pushmeet Kohli, Marcello Pelillo, Horst Bischof

2 0.70315665 232 nips-2012-Multiplicative Forests for Continuous-Time Processes

Author: Jeremy Weiss, Sriraam Natarajan, David Page

3 0.68543601 339 nips-2012-The Time-Marginalized Coalescent Prior for Hierarchical Clustering

Author: Levi Boyles, Max Welling

4 0.6686396 207 nips-2012-Mandatory Leaf Node Prediction in Hierarchical Multilabel Classification

Author: Wei Bi, James T. Kwok

Abstract: In hierarchical classiﬁcation, the prediction paths may be required to always end at leaf nodes. This is called mandatory leaf node prediction (MLNP) and is particularly useful when the leaf nodes have much stronger semantic meaning than the internal nodes. However, while there have been a lot of MLNP methods in hierarchical multiclass classiﬁcation, performing MLNP in hierarchical multilabel classiﬁcation is much more difﬁcult. In this paper, we propose a novel MLNP algorithm that (i) considers the global hierarchy structure; and (ii) can be used on hierarchies of both trees and DAGs. We show that one can efﬁciently maximize the joint posterior probability of all the node labels by a simple greedy algorithm. Moreover, this can be further extended to the minimization of the expected symmetric loss. Experiments are performed on a number of real-world data sets with tree- and DAG-structured label hierarchies. The proposed method consistently outperforms other hierarchical and ﬂat multilabel classiﬁcation methods. 1

5 0.64519745 183 nips-2012-Learning Partially Observable Models Using Temporally Abstract Decision Trees

Author: Erik Talvitie

Abstract: This paper introduces timeline trees, which are partial models of partially observable environments. Timeline trees are given some speciﬁc predictions to make and learn a decision tree over history. The main idea of timeline trees is to use temporally abstract features to identify and split on features of key events, spread arbitrarily far apart in the past (whereas previous decision-tree-based methods have been limited to a ﬁnite sufﬁx of history). Experiments demonstrate that timeline trees can learn to make high quality predictions in complex, partially observable environments with high-dimensional observations (e.g. an arcade game). 1

6 0.59980619 215 nips-2012-Minimizing Uncertainty in Pipelines

7 0.57916695 306 nips-2012-Semantic Kernel Forests from Multiple Taxonomies

8 0.5783596 260 nips-2012-Online Sum-Product Computation Over Trees

9 0.57620823 40 nips-2012-Analyzing 3D Objects in Cluttered Images

10 0.57613379 78 nips-2012-Compressive Sensing MRI with Wavelet Tree Sparsity

11 0.54236215 357 nips-2012-Unsupervised Template Learning for Fine-Grained Object Recognition

12 0.52858126 344 nips-2012-Timely Object Recognition

13 0.52804613 106 nips-2012-Dynamical And-Or Graph Learning for Object Shape Modeling and Detection

14 0.52711856 8 nips-2012-A Generative Model for Parts-based Object Segmentation

15 0.52355933 201 nips-2012-Localizing 3D cuboids in single-view images

16 0.5190571 58 nips-2012-Bayesian models for Large-scale Hierarchical Classification

17 0.50793809 87 nips-2012-Convolutional-Recursive Deep Learning for 3D Object Classification

18 0.50561965 101 nips-2012-Discriminatively Trained Sparse Code Gradients for Contour Detection

19 0.49199995 209 nips-2012-Max-Margin Structured Output Regression for Spatio-Temporal Action Localization

20 0.4843919 210 nips-2012-Memorability of Image Regions

similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(0, 0.035), (17, 0.011), (21, 0.032), (38, 0.092), (39, 0.019), (42, 0.02), (44, 0.29), (54, 0.036), (55, 0.015), (74, 0.124), (76, 0.138), (80, 0.08), (92, 0.034)]

similar papers list:

simIndex simValue paperId paperTitle

1 0.78740621 193 nips-2012-Learning to Align from Scratch

Author: Gary Huang, Marwan Mattar, Honglak Lee, Erik G. Learned-miller

Abstract: Unsupervised joint alignment of images has been demonstrated to improve performance on recognition tasks such as face veriﬁcation. Such alignment reduces undesired variability due to factors such as pose, while only requiring weak supervision in the form of poorly aligned examples. However, prior work on unsupervised alignment of complex, real-world images has required the careful selection of feature representation based on hand-crafted image descriptors, in order to achieve an appropriate, smooth optimization landscape. In this paper, we instead propose a novel combination of unsupervised joint alignment with unsupervised feature learning. Speciﬁcally, we incorporate deep learning into the congealing alignment framework. Through deep learning, we obtain features that can represent the image at differing resolutions based on network depth, and that are tuned to the statistics of the speciﬁc data being aligned. In addition, we modify the learning algorithm for the restricted Boltzmann machine by incorporating a group sparsity penalty, leading to a topographic organization of the learned ﬁlters and improving subsequent alignment results. We apply our method to the Labeled Faces in the Wild database (LFW). Using the aligned images produced by our proposed unsupervised algorithm, we achieve higher accuracy in face veriﬁcation compared to prior work in both unsupervised and supervised alignment. We also match the accuracy for the best available commercial method. 1

same-paper 2 0.77566952 81 nips-2012-Context-Sensitive Decision Forests for Object Detection

Author: Peter Kontschieder, Samuel R. Bulò, Antonio Criminisi, Pushmeet Kohli, Marcello Pelillo, Horst Bischof

3 0.77386451 150 nips-2012-Hierarchical spike coding of sound

Author: Yan Karklin, Chaitanya Ekanadham, Eero P. Simoncelli

Abstract: Natural sounds exhibit complex statistical regularities at multiple scales. Acoustic events underlying speech, for example, are characterized by precise temporal and frequency relationships, but they can also vary substantially according to the pitch, duration, and other high-level properties of speech production. Learning this structure from data while capturing the inherent variability is an important ﬁrst step in building auditory processing systems, as well as understanding the mechanisms of auditory perception. Here we develop Hierarchical Spike Coding, a two-layer probabilistic generative model for complex acoustic structure. The ﬁrst layer consists of a sparse spiking representation that encodes the sound using kernels positioned precisely in time and frequency. Patterns in the positions of ﬁrst layer spikes are learned from the data: on a coarse scale, statistical regularities are encoded by a second-layer spiking representation, while ﬁne-scale structure is captured by recurrent interactions within the ﬁrst layer. When ﬁt to speech data, the second layer acoustic features include harmonic stacks, sweeps, frequency modulations, and precise temporal onsets, which can be composed to represent complex acoustic events. Unlike spectrogram-based methods, the model gives a probability distribution over sound pressure waveforms. This allows us to use the second-layer representation to synthesize sounds directly, and to perform model-based denoising, on which we demonstrate a signiﬁcant improvement over standard methods. 1

4 0.68621498 148 nips-2012-Hamming Distance Metric Learning

Author: Mohammad Norouzi, David M. Blei, Ruslan Salakhutdinov

Abstract: Motivated by large-scale multimedia applications we propose to learn mappings from high-dimensional data to binary codes that preserve semantic similarity. Binary codes are well suited to large-scale applications as they are storage efﬁcient and permit exact sub-linear kNN search. The framework is applicable to broad families of mappings, and uses a ﬂexible form of triplet ranking loss. We overcome discontinuous optimization of the discrete mappings by minimizing a piecewise-smooth upper bound on empirical loss, inspired by latent structural SVMs. We develop a new loss-augmented inference algorithm that is quadratic in the code length. We show strong retrieval performance on CIFAR-10 and MNIST, with promising classiﬁcation results using no more than kNN on the binary codes. 1

5 0.64232057 199 nips-2012-Link Prediction in Graphs with Autoregressive Features

Author: Emile Richard, Stephane Gaiffas, Nicolas Vayatis

Abstract: In the paper, we consider the problem of link prediction in time-evolving graphs. We assume that certain graph features, such as the node degree, follow a vector autoregressive (VAR) model and we propose to use this information to improve the accuracy of prediction. Our strategy involves a joint optimization procedure over the space of adjacency matrices and VAR matrices which takes into account both sparsity and low rank properties of the matrices. Oracle inequalities are derived and illustrate the trade-offs in the choice of smoothing parameters when modeling the joint effect of sparsity and low rank property. The estimate is computed efﬁciently using proximal methods through a generalized forward-backward agorithm. 1

6 0.61877161 3 nips-2012-A Bayesian Approach for Policy Learning from Trajectory Preference Queries

7 0.61456209 274 nips-2012-Priors for Diversity in Generative Latent Variable Models

8 0.60932714 339 nips-2012-The Time-Marginalized Coalescent Prior for Hierarchical Clustering

9 0.60598201 210 nips-2012-Memorability of Image Regions

10 0.60398877 201 nips-2012-Localizing 3D cuboids in single-view images

11 0.60229516 176 nips-2012-Learning Image Descriptors with the Boosting-Trick

12 0.60228157 357 nips-2012-Unsupervised Template Learning for Fine-Grained Object Recognition

13 0.59963155 303 nips-2012-Searching for objects driven by context

14 0.59707683 168 nips-2012-Kernel Latent SVM for Visual Recognition

15 0.59665012 8 nips-2012-A Generative Model for Parts-based Object Segmentation

16 0.59547091 209 nips-2012-Max-Margin Structured Output Regression for Spatio-Temporal Action Localization

17 0.59303814 185 nips-2012-Learning about Canonical Views from Internet Image Collections

18 0.59119219 337 nips-2012-The Lovász ϑ function, SVMs and finding large dense subgraphs

19 0.58900648 101 nips-2012-Discriminatively Trained Sparse Code Gradients for Contour Detection

20 0.58842999 163 nips-2012-Isotropic Hashing