nips nips2012 nips2012-8 knowledge-graph by maker-knowledge-mining

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

Source: pdf

Author: S. Eslami, Christopher Williams

Abstract: The Shape Boltzmann Machine (SBM) [1] has recently been introduced as a stateof-the-art model of foreground/background object shape. We extend the SBM to account for the foreground object’s parts. Our new model, the Multinomial SBM (MSBM), can capture both local and global statistics of part shapes accurately. We combine the MSBM with an appearance model to form a fully generative model of images of objects. Parts-based object segmentations are obtained simply by performing probabilistic inference in the model. We apply the model to two challenging datasets which exhibit signiﬁcant shape and appearance variability, and ﬁnd that it obtains results that are comparable to the state-of-the-art. There has been signiﬁcant focus in computer vision on object recognition and detection e.g. [2], but a strong desire remains to obtain richer descriptions of objects than just their bounding boxes. One such description is a parts-based object segmentation, in which an image is partitioned into multiple sets of pixels, each belonging to either a part of the object of interest, or its background. The signiﬁcance of parts in computer vision has been recognized since the earliest days of the ﬁeld (e.g. [3, 4, 5]), and there exists a rich history of work on probabilistic models for parts-based segmentation e.g. [6, 7]. Many such models only consider local neighborhood statistics, however several models have recently been proposed that aim to increase the accuracy of segmentations by also incorporating prior knowledge about the foreground object’s shape [8, 9, 10, 11]. In such cases, probabilistic techniques often mainly differ in how accurately they represent and learn about the variability exhibited by the shapes of the object’s parts. Accurate models of the shapes and appearances of parts can be necessary to perform inference in datasets that exhibit large amounts of variability. In general, the stronger the models of these two components, the more performance is improved. A generative model has the added beneﬁt of being able to generate samples, which allows us to visually inspect the quality of its understanding of the data and the problem. Recently, a generative probabilistic model known as the Shape Boltzmann Machine (SBM) has been used to model binary object shapes [1]. The SBM has been shown to constitute the state-of-the-art and it possesses several highly desirable characteristics: samples from the model look realistic, and it generalizes to generate samples that differ from the limited number of examples it is trained on. The main contributions of this paper are as follows: 1) In order to account for object parts we extend the SBM to use multinomial visible units instead of binary ones, resulting in the Multinomial Shape Boltzmann Machine (MSBM), and we demonstrate that the MSBM constitutes a strong model of parts-based object shape. 2) We combine the MSBM with an appearance model to form a fully generative model of images of objects (see Fig. 1). We show how parts-based object segmentations can be obtained simply by performing probabilistic inference in the model. We apply our model to two challenging datasets and ﬁnd that in addition to being principled and fully generative, the model’s performance is comparable to the state-of-the-art. 1 Train labels Train images Test image Appearance model Joint Model Shape model Parsing Figure 1: Overview. Using annotated images separate models of shape and appearance are trained. Given an unseen test image, its parsing is obtained via inference in the proposed joint model. In Secs. 1 and 2 we present the model and propose efﬁcient inference and learning schemes. In Sec. 3 we compare and contrast the resulting joint model with existing work in the literature. We describe our experimental results in Sec. 4 and conclude with a discussion in Sec. 5. 1 Model We consider datasets of cropped images of an object class. We assume that the images are constructed through some combination of a ﬁxed number of parts. Given a dataset D = {Xd }, d = 1...n of such images X, each consisting of P pixels {xi }, i = 1...P , we wish to infer a segmentation S for the image. S consists of a labeling si for every pixel, where si is a 1-of-(L+1) encoded variable, and L is the ﬁxed number of parts that combine to generate the foreground. In other words, si = (sli ), P l = 0...L, sli 2 {0, 1} and l sli = 1. Note that the background is also treated as a ‘part’ (l = 0). Accurate inference of S is driven by models for 1) part shapes and 2) part appearances. Part shapes: Several types of models can be used to deﬁne probabilistic distributions over segmentations S. The simplest approach is to model each pixel si independently with categorical variables whose parameters are speciﬁed by the object’s mean shape (Fig. 2(a)). Markov Random Fields (MRFs, Fig. 2(b)) additionally model interactions between nearby pixels using pairwise potential functions that efﬁciently capture local properties of images like smoothness and continuity. Restricted Boltzmann Machines (RBMs) and their multi-layered counterparts Deep Boltzmann Machines (DBMs, Fig. 2(c)) make heavy use of hidden variables to efﬁciently deﬁne higher-order potentials that take into account the conﬁguration of larger groups of image pixels. The introduction of such hidden variables provides a way to efﬁciently capture complex, global properties of image pixels. RBMs and DBMs are powerful generative models, but they also have many parameters. Segmented images, however, are expensive to obtain and datasets are typically small (hundreds of examples). In order to learn a model that accurately captures the properties of part shapes we use DBMs but also impose carefully chosen connectivity and capacity constraints, following the structure of the Shape Boltzmann Machine (SBM) [1]. We further extend the model to account for multi-part shapes to obtain the Multinomial Shape Boltzmann Machine (MSBM). The MSBM has two layers of latent variables: h1 and h2 (collectively H = {h1 , h2 }), and deﬁnes a P Boltzmann distribution over segmentations p(S) = h1 ,h2 exp{ E(S, h1 , h2 |✓s )}/Z(✓s ) where X X X X X 1 2 E(S, h1 , h2 |✓s ) = bli sli + wlij sli h1 + c 1 h1 + wjk h1 h2 + c2 h2 , (1) j j j j k k k i,l j i,j,l j,k k where j and k range over the ﬁrst and second layer hidden variables, and ✓s = {W 1 , W 2 , b, c1 , c2 } are the shape model parameters. In the ﬁrst layer, local receptive ﬁelds are enforced by connecting each hidden unit in h1 only to a subset of the visible units, corresponding to one of four patches, as shown in Fig. 2(d,e). Each patch overlaps its neighbor by b pixels, which allows boundary continuity to be learned at the lowest layer. We share weights between the four sets of ﬁrst-layer hidden units and patches, and purposely restrict the number of units in h2 . These modiﬁcations signiﬁcantly reduce the number of parameters whilst taking into account an important property of shapes, namely that the strongest dependencies between pixels are typically local. 2 h2 1 1 h S S (a) Mean h S (b) MRF h2 h2 h1 S S (c) DBM b (d) SBM (e) 2D SBM Figure 2: Models of shape. Object shape is modeled with undirected graphical models. (a) 1D slice of a mean model. (b) Markov Random Field in 1D. (c) Deep Boltzmann Machine in 1D. (d) 1D slice of a Shape Boltzmann Machine. (e) Shape Boltzmann Machine in 2D. In all models latent units h are binary and visible units S are multinomial random variables. Based on Fig. 2 of [1]. k=1 k=2 k=3 k=1 k=2 k=3 k=1 k=2 k=3 ⇡ l=0 l=1 l=2 Figure 3: A model of appearances. Left: An exemplar dataset. Here we assume one background (l = 0) and two foreground (l = 1, non-body; l = 2, body) parts. Right: The corresponding appearance model. In this example, L = 2, K = 3 and W = 6. Best viewed in color. Part appearances: Pixels in a given image are assumed to have been generated by W ﬁxed Gaussians in RGB space. During pre-training, the means {µw } and covariances {⌃w } of these Gaussians are extracted by training a mixture model with W components on every pixel in the dataset, ignoring image and part structure. It is also assumed that each of the L parts can have different appearances in different images, and that these appearances can be clustered into K classes. The classes differ in how likely they are to use each of the W components when ‘coloring in’ the part. The generative process is as follows. For part l in an image, one of the K classes is chosen (represented by a 1-of-K indicator variable al ). Given al , the probability distribution deﬁned on pixels associated with part l is given by a Gaussian mixture model with means {µw } and covariances {⌃w } and mixing proportions { lkw }. The prior on A = {al } speciﬁes the probability ⇡lk of appearance class k being chosen for part l. Therefore appearance parameters ✓a = {⇡lk , lkw } (see Fig. 3) and: a p(xi |A, si , ✓ ) = p(A|✓a ) = Y l Y l a sli p(xi |al , ✓ ) p(al |✓a ) = = Y Y X YY l l k w lkw N (xi |µw , ⌃w ) !alk !sli (⇡lk )alk . , (2) (3) k Combining shapes and appearances: To summarize, the latent variables for X are A, S, H, and the model’s active parameters ✓ include shape parameters ✓s and appearance parameters ✓a , so that p(X, A, S, H|✓) = Y 1 p(A|✓a )p(S, H|✓s ) p(xi |A, si , ✓a ) , Z( ) i (4) where the parameter adjusts the relative contributions of the shape and appearance components. See Fig. 4 for an illustration of the complete graphical model. During learning, we ﬁnd the values of ✓ that maximize the likelihood of the training data D, and segmentation is performed on a previously-unseen image by querying the marginal distribution p(S|Xtest , ✓). Note that Z( ) is constant throughout the execution of the algorithms. We set via trial and error in our experiments. 3 n H ✓a si al H xi L+1 ✓s S X A P Figure 4: A model of shape and appearance. Left: The joint model. Pixels xi are modeled via appearance variables al . The model’s belief about each layer’s shape is captured by shape variables H. Segmentation variables si assign each pixel to a layer. Right: Schematic for an image X. 2 Inference and learning Inference: We approximate p(A, S, H|X, ✓) by drawing samples of A, S and H using block-Gibbs Markov Chain Monte Carlo (MCMC). The desired distribution p(S|X, ✓) can then be obtained by considering only the samples for S (see Algorithm 1). In order to sample p(A|S, H, X, ✓) we consider the conditional distribution of appearance class k being chosen for part l which is given by: Q P ·s ⇡lk i ( w lkw N (xi |µw , ⌃w )) li h Q P i. p(alk = 1|S, X, ✓) = P (5) K ·sli r=1 ⇡lr i( w lrw N (xi |µw , ⌃w )) Since the MSBM only has edges between each pair of adjacent layers, all hidden units within a layer are conditionally independent given the units in the other two layers. This property can be exploited to make inference in the shape model exact and efﬁcient. The conditional probabilities are: X X 1 2 p(h1 = 1|s, h2 , ✓) = ( wlij sli + wjk h2 + c1 ), (6) j k j i,l p(h2 k 1 = 1|h , ✓) = ( X k 2 wjk h1 j + c2 ), j (7) j where (y) = 1/(1 + exp( y)) is the sigmoid function. To sample from p(H|S, X, ✓) we iterate between Eqns. 6 and 7 multiple times and keep only the ﬁnal values of h1 and h2 . Finally, we draw samples for the pixels in p(S|A, H, X, ✓) independently: P 1 exp( j wlij h1 + bli ) p(xi |A, sli = 1, ✓) j p(sli = 1|A, H, X, ✓) = PL . (8) P 1 1 m=1 exp( j wmij hj + bmi ) p(xi |A, smi = 1, ✓) Seeding: Since the latent-space is extremely high-dimensional, in practice we ﬁnd it helpful to run several inference chains, each initializing S(1) to a different value. The ‘best’ inference is retained and the others are discarded. The computation of the likelihood p(X|✓) of image X is intractable, so we approximate the quality of each inference using a scoring function: 1X Score(X|✓) = p(X, A(t) , S(t) , H(t) |✓), (9) T t where {A(t) , S(t) , H(t) }, t = 1...T are the samples obtained from the posterior p(A, S, H|X, ✓). If the samples were drawn from the prior p(A, S, H|✓) the scoring function would be an unbiased estimator of p(X|✓), but would be wildly inaccurate due to the high probability of missing the important regions of latent space (see e.g. [12, p. 107-109] for further discussion of this issue). Learning: Learning of the model involves maximizing the log likelihood log p(D|✓a , ✓s ) of the training dataset D with respect to the model parameters ✓a and ✓s . Since training is partially supervised, in that for each image X its corresponding segmentation S is also given, we can learn the parameters of the shape and appearance components separately. For appearances, the learning of the mixing coefﬁcients and the histogram parameters decomposes into standard mixture updates independently for each part. For shapes, we follow the standard deep 4 Algorithm 1 MCMC inference algorithm. 1: procedure I NFER(X, ✓) 2: Initialize S(1) , H(1) 3: for t 2 : chain length do 4: A(t) ⇠ p(A|S(t 1) , H(t 1) , X, ✓) 5: S(t) ⇠ p(S|A(t) , H(t 1) , X, ✓) 6: H(t) ⇠ p(H|S(t) , ✓) 7: return {S(t) }t=burnin:chain length learning literature closely [13, 1]. In the pre-training phase we greedily train the model bottom up, one layer at a time. We begin by training an RBM on the observed data using stochastic maximum likelihood learning (SML; also referred to as ‘persistent CD’; [14, 13]). Once this RBM is trained, we infer the conditional mean of the hidden units for each training image. The resulting vectors then serve as the training data for a second RBM which is again trained using SML. We use the parameters of these two RBMs to initialize the parameters of the full MSBM model. In the second phase we perform approximate stochastic gradient ascent in the likelihood of the full model to ﬁnetune the parameters in an EM-like scheme as described in [13]. 3 Related work Existing probabilistic models of images can be categorized by the amount of variability they expect to encounter in the data and by how they model this variability. A signiﬁcant portion of the literature models images using only two parts: a foreground object and its background e.g. [15, 16, 17, 18, 19]. Models that account for the parts within the foreground object mainly differ in how accurately they learn about and represent the variability of the shapes of the object’s parts. In Probabilistic Index Maps (PIMs) [8] a mean partitioning is learned, and the deformable PIM [9] additionally allows for local deformations of this mean partitioning. Stel Component Analysis [10] accounts for larger amounts of shape variability by learning a number of different template means for the object that are blended together on a pixel-by-pixel basis. Factored Shapes and Appearances [11] models global properties of shape using a factor analysis-like model, and ‘masked’ RBMs have been used to model more local properties of shape [20]. However, none of these models constitute a strong model of shape in terms of realism of samples and generalization capabilities [1]. We demonstrate in Sec. 4 that, like the SBM, the MSBM does in fact possess these properties. The closest works to ours in terms of ability to deal with datasets that exhibit signiﬁcant variability in both shape and appearance are the works of Bo and Fowlkes [21] and Thomas et al. [22]. Bo and Fowlkes [21] present an algorithm for pedestrian segmentation that models the shapes of the parts using several template means. The different parts are composed using hand coded geometric constraints, which means that the model cannot be automatically extended to other application domains. The Implicit Shape Model (ISM) used in [22] is reliant on interest point detectors and deﬁnes distributions over segmentations only in the posterior, and therefore is not fully generative. The model presented here is entirely learned from data and fully generative, therefore it can be applied to new datasets and diagnosed with relative ease. Due to its modular structure, we also expect it to rapidly absorb future developments in shape and appearance models. 4 Experiments Penn-Fudan pedestrians: The ﬁrst dataset that we considered is Penn-Fudan pedestrians [23], consisting of 169 images of pedestrians (Fig. 6(a)). The images are annotated with ground-truth segmentations for L = 7 different parts (hair, face, upper and lower clothes, shoes, legs, arms; Fig. 6(d)). We compare the performance of the model with the algorithm of Bo and Fowlkes [21]. For the shape component, we trained an MSBM on the 684 images of a labeled version of the HumanEva dataset [24] (at 48 ⇥ 24 pixels; also ﬂipped horizontally) with overlap b = 4, and 400 and 50 hidden units in the ﬁrst and second layers respectively. Each layer was pre-trained for 3000 epochs (iterations). After pre-training, joint training was performed for 1000 epochs. 5 (c) Completion (a) Sampling (b) Diffs ! ! ! Figure 5: Learned shape model. (a) A chain of samples (1000 samples between frames). The apparent ‘blurriness’ of samples is not due to averaging or resizing. We display the probability of each pixel belonging to different parts. If, for example, there is a 50-50 chance that a pixel belongs to the red or blue parts, we display that pixel in purple. (b) Differences between the samples and their most similar counterparts in the training dataset. (c) Completion of occlusions (pink). To assess the realism and generalization characteristics of the learned MSBM we sample from it. In Fig. 5(a) we show a chain of unconstrained samples from an MSBM generated via block-Gibbs MCMC (1000 samples between frames). The model captures highly non-linear correlations in the data whilst preserving the object’s details (e.g. face and arms). To demonstrate that the model has not simply memorized the training data, in Fig. 5(b) we show the difference between the sampled shapes in Fig. 5(a) and their closest images in the training set (based on per-pixel label agreement). We see that the model generalizes in non-trivial ways to generate realistic shapes that it had not encountered during training. In Fig. 5(c) we show how the MSBM completes rectangular occlusions. The samples highlight the variability in possible completions captured by the model. Note how, e.g. the length of the person’s trousers on one leg affects the model’s predictions for the other, demonstrating the model’s knowledge about long-range dependencies. An interactive M ATLAB GUI for sampling from this MSBM has been included in the supplementary material. The Penn-Fudan dataset (at 200 ⇥ 100 pixels) was then split into 10 train/test cross-validation splits without replacement. We used the training images in each split to train the appearance component with a vocabulary of size W = 50 and K = 100 mixture components1 . We additionally constrained the model by sharing the appearance models for the arms and legs with that of the face. We assess the quality of the appearance model by performing the following experiment: for each test image, we used the scoring function described in Eq. 9 to evaluate a number of different proposal segmentations for that image. We considered 10 randomly chosen segmentations from the training dataset as well as the ground-truth segmentation for the test image, and found that the appearance model correctly assigns the highest score to the ground-truth 95% of the time. During inference, the shape and appearance models (which are deﬁned on images of different sizes), were combined at 200 ⇥ 100 pixels via M ATLAB’s imresize function, and we set = 0.8 (Eq. 8) via trial and error. Inference chains were seeded at 100 exemplar segmentations from the HumanEva dataset (obtained using the K-medoids algorithm with K = 100), and were run for 20 Gibbs iterations each (with 5 iterations of Eqs. 6 and 7 per Gibbs iteration). Our unoptimized M ATLAB implementation completed inference for each chain in around 7 seconds. We compute the conditional probability of each pixel belonging to different parts given the last set of samples obtained from the highest scoring chain, assign each pixel independently to the most likely part at that pixel, and report the percentage of correctly labeled pixels (see Table 1). We ﬁnd that accuracy can be improved using superpixels (SP) computed on X (pixels within a superpixel are all assigned the most common label within it; as with [21] we use gPb-OWT-UCM [25]). We also report the accuracy obtained, had the top scoring seed segmentation been returned as the ﬁnal segmentation for each image. Here the quality of the seed is determined solely by the appearance model. We observe that the model has comparable performance to the state-of-the-art but pedestrianspeciﬁc algorithm of [21], and that inference in the model signiﬁcantly improves the accuracy of the segmentations over the baseline (top seed+SP). Qualitative results can be seen in Fig. 6(c). 1 We obtained the best quantitative results with these settings. The appearances exhibited by the parts in the dataset are highly varied, and the complexity of the appearance model reﬂects this fact. 6 Table 1: Penn-Fudan pedestrians. We report the percentage of correctly labeled pixels. The ﬁnal column is an average of the background, upper and lower body scores (as reported in [21]). FG BG Upper Body Lower Body Head Average Bo and Fowlkes [21] 73.3% 81.1% 73.6% 71.6% 51.8% 69.5% MSBM MSBM + SP 70.7% 71.6% 72.8% 73.8% 68.6% 69.9% 66.7% 68.5% 53.0% 54.1% 65.3% 66.6% Top seed Top seed + SP 59.0% 61.6% 61.8% 67.3% 56.8% 60.8% 49.8% 54.1% 45.5% 43.5% 53.5% 56.4% Table 2: ETHZ cars. We report the percentage of pixels belonging to each part that are labeled correctly. The ﬁnal column is an average weighted by the frequency of occurrence of each label. BG Body Wheel Window Bumper License Light Average ISM [22] 93.2% 72.2% 63.6% 80.5% 73.8% 56.2% 34.8% 86.8% MSBM 94.6% 72.7% 36.8% 74.4% 64.9% 17.9% 19.9% 86.0% Top seed 92.2% 68.4% 28.3% 63.8% 45.4% 11.2% 15.1% 81.8% ETHZ cars: The second dataset that we considered is the ETHZ labeled cars dataset [22], which itself is a subset of the LabelMe dataset [23], consisting of 139 images of cars, all in the same semiproﬁle view (Fig. 7(a)). The images are annotated with ground-truth segmentations for L = 6 parts (body, wheel, window, bumper, license plate, headlight; Fig. 7(d)). We compare the performance of the model with the ISM of Thomas et al. [22], who also report their results on this dataset. The dataset was split into 10 train/test cross-validation splits without replacement. We used the training images in each split to train both the shape and appearance components. For the shape component, we trained an MSBM at 50 ⇥ 50 pixels with overlap b = 4, and 2000 and 100 hidden units in the ﬁrst and second layers respectively. Each layer was pre-trained for 3000 epochs and joint training was performed for 1000 epochs. The appearance model was trained with a vocabulary of size W = 50 and K = 100 mixture components and we set = 0.7. Inference chains were seeded at 50 exemplar segmentations (obtained using K-medoids). We ﬁnd that the use of superpixels does not help with this dataset (due to the poor quality of superpixels obtained for these images). Qualitative and quantitative results that show the performance of model to be comparable to the state-of-the-art ISM can be seen in Fig. 7(c) and Table 2. We believe the discrepancy in accuracy between the MSBM and ISM on the ‘license’ and ‘light’ labels to mainly be due to ISM’s use of interest-points, as they are able to locate such ﬁne structures accurately. By incorporating better models of part appearance into the generative model, we expect to see this discrepancy decrease. 5 Conclusions and future work In this paper we have shown how the SBM can be extended to obtain the MSBM, and presented a principled probabilistic model of images of objects that exploits the MSBM as its model for part shapes. We demonstrated how object segmentations can be obtained simply by performing MCMC inference in the model. The model can also be treated as a probabilistic evaluator of segmentations: given a proposal segmentation it can be used to estimate its likelihood. This leads us to believe that the combination of a generative model such as ours, with a discriminative, bottom-up segmentation algorithm could be highly effective. We are currently investigating how textured appearance models, which take into account the spatial structure of pixels, affect the learning and inference algorithms and the performance of the model. Acknowledgments Thanks to Charless Fowlkes and Vittorio Ferrari for access to datasets, and to Pushmeet Kohli and John Winn for valuable discussions. AE has received funding from the Carnegie Trust, the SORSAS scheme, and the IST Programme under the PASCAL2 Network of Excellence (IST-2007-216886). 7 (a) Test (c) MSBM (b) Bo and Fowlkes (d) Ground truth Background Hair Face Upper Shoes Legs Lower Arms (d) Ground truth (c) MSBM (b) Thomas et al. (a) Test Figure 6: Penn-Fudan pedestrians. (a) Test images. (b) Results reported by Bo and Fowlkes [21]. (c) Output of the joint model. (d) Ground-truth images. Images shown are those selected by [21]. Background Body Wheel Window Bumper License Headlight Figure 7: ETHZ cars. (a) Test images. (b) Results reported by Thomas et al. [22]. (c) Output of the joint model. (d) Ground-truth images. Images shown are those selected by [22]. 8 References [1] S. M. Ali Eslami, Nicolas Heess, and John Winn. The Shape Boltzmann Machine: a Strong Model of Object Shape. In IEEE CVPR, 2012. [2] Mark Everingham, Luc Van Gool, Christopher K. I. Williams, John Winn, and Andrew Zisserman. The PASCAL Visual Object Classes (VOC) Challenge. International Journal of Computer Vision, 88:303–338, 2010. [3] Martin Fischler and Robert Elschlager. The Representation and Matching of Pictorial Structures. IEEE Transactions on Computers, 22(1):67–92, 1973. [4] David Marr. Vision: A Computational Investigation into the Human Representation and Processing of Visual Information. Freeman, 1982. [5] Irving Biederman. Recognition-by-components: A theory of human image understanding. Psychological Review, 94:115–147, 1987. [6] Ashish Kapoor and John Winn. Located Hidden Random Fields: Learning Discriminative Parts for Object Detection. In ECCV, pages 302–315, 2006. [7] John Winn and Jamie Shotton. The Layout Consistent Random Field for Recognizing and Segmenting Partially Occluded Objects. In IEEE CVPR, pages 37–44, 2006. [8] Nebojsa Jojic and Yaron Caspi. Capturing Image Structure with Probabilistic Index Maps. In IEEE CVPR, pages 212–219, 2004. [9] John Winn and Nebojsa Jojic. LOCUS: Learning object classes with unsupervised segmentation. In ICCV, pages 756–763, 2005. [10] Nebojsa Jojic, Alessandro Perina, Marco Cristani, Vittorio Murino, and Brendan Frey. Stel component analysis. In IEEE CVPR, pages 2044–2051, 2009. [11] S. M. Ali Eslami and Christopher K. I. Williams. Factored Shapes and Appearances for Partsbased Object Understanding. In BMVC, pages 18.1–18.12, 2011. [12] Nicolas Heess. Learning generative models of mid-level structure in natural images. PhD thesis, University of Edinburgh, 2011. [13] Ruslan Salakhutdinov and Geoffrey Hinton. Deep Boltzmann Machines. In AISTATS, volume 5, pages 448–455, 2009. [14] Tijmen Tieleman. Training restricted Boltzmann machines using approximations to the likelihood gradient. In ICML, pages 1064–1071, 2008. [15] Carsten Rother, Vladimir Kolmogorov, and Andrew Blake. “GrabCut”: interactive foreground extraction using iterated graph cuts. ACM SIGGRAPH, 23:309–314, 2004. [16] Eran Borenstein, Eitan Sharon, and Shimon Ullman. Combining Top-Down and Bottom-Up Segmentation. In CVPR Workshop on Perceptual Organization in Computer Vision, 2004. [17] Himanshu Arora, Nicolas Loeff, David Forsyth, and Narendra Ahuja. Unsupervised Segmentation of Objects using Efﬁcient Learning. IEEE CVPR, pages 1–7, 2007. [18] Bogdan Alexe, Thomas Deselaers, and Vittorio Ferrari. ClassCut for unsupervised class segmentation. In ECCV, pages 380–393, 2010. [19] Nicolas Heess, Nicolas Le Roux, and John Winn. Weakly Supervised Learning of ForegroundBackground Segmentation using Masked RBMs. In ICANN, 2011. [20] Nicolas Le Roux, Nicolas Heess, Jamie Shotton, and John Winn. Learning a Generative Model of Images by Factoring Appearance and Shape. Neural Computation, 23(3):593–650, 2011. [21] Yihang Bo and Charless Fowlkes. Shape-based Pedestrian Parsing. In IEEE CVPR, 2011. [22] Alexander Thomas, Vittorio Ferrari, Bastian Leibe, Tinne Tuytelaars, and Luc Van Gool. Using Recognition and Annotation to Guide a Robot’s Attention. IJRR, 28(8):976–998, 2009. [23] Bryan Russell, Antonio Torralba, Kevin Murphy, and William Freeman. LabelMe: A Database and Tool for Image Annotation. International Journal of Computer Vision, 77:157–173, 2008. [24] Leonid Sigal, Alexandru Balan, and Michael Black. HumanEva. International Journal of Computer Vision, 87(1-2):4–27, 2010. [25] Pablo Arbelaez, Michael Maire, Charless C. Fowlkes, and Jitendra Malik. From Contours to Regions: An Empirical Evaluation. In IEEE CVPR, 2009. 9

Reference: text

Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 uk Abstract The Shape Boltzmann Machine (SBM) [1] has recently been introduced as a stateof-the-art model of foreground/background object shape. [sent-13, score-0.156]

2 We extend the SBM to account for the foreground object’s parts. [sent-14, score-0.127]

3 Our new model, the Multinomial SBM (MSBM), can capture both local and global statistics of part shapes accurately. [sent-15, score-0.214]

4 We combine the MSBM with an appearance model to form a fully generative model of images of objects. [sent-16, score-0.473]

5 Parts-based object segmentations are obtained simply by performing probabilistic inference in the model. [sent-17, score-0.434]

6 We apply the model to two challenging datasets which exhibit signiﬁcant shape and appearance variability, and ﬁnd that it obtains results that are comparable to the state-of-the-art. [sent-18, score-0.463]

7 There has been signiﬁcant focus in computer vision on object recognition and detection e. [sent-19, score-0.159]

8 One such description is a parts-based object segmentation, in which an image is partitioned into multiple sets of pixels, each belonging to either a part of the object of interest, or its background. [sent-22, score-0.393]

9 The signiﬁcance of parts in computer vision has been recognized since the earliest days of the ﬁeld (e. [sent-23, score-0.121]

10 [3, 4, 5]), and there exists a rich history of work on probabilistic models for parts-based segmentation e. [sent-25, score-0.169]

11 Many such models only consider local neighborhood statistics, however several models have recently been proposed that aim to increase the accuracy of segmentations by also incorporating prior knowledge about the foreground object’s shape [8, 9, 10, 11]. [sent-28, score-0.524]

12 In such cases, probabilistic techniques often mainly differ in how accurately they represent and learn about the variability exhibited by the shapes of the object’s parts. [sent-29, score-0.267]

13 Accurate models of the shapes and appearances of parts can be necessary to perform inference in datasets that exhibit large amounts of variability. [sent-30, score-0.455]

14 Recently, a generative probabilistic model known as the Shape Boltzmann Machine (SBM) has been used to model binary object shapes [1]. [sent-33, score-0.46]

15 The SBM has been shown to constitute the state-of-the-art and it possesses several highly desirable characteristics: samples from the model look realistic, and it generalizes to generate samples that differ from the limited number of examples it is trained on. [sent-34, score-0.136]

16 2) We combine the MSBM with an appearance model to form a fully generative model of images of objects (see Fig. [sent-36, score-0.473]

17 We show how parts-based object segmentations can be obtained simply by performing probabilistic inference in the model. [sent-38, score-0.434]

18 1 Train labels Train images Test image Appearance model Joint Model Shape model Parsing Figure 1: Overview. [sent-40, score-0.237]

19 Using annotated images separate models of shape and appearance are trained. [sent-41, score-0.555]

20 1 Model We consider datasets of cropped images of an object class. [sent-50, score-0.248]

21 We assume that the images are constructed through some combination of a ﬁxed number of parts. [sent-51, score-0.12]

22 n of such images X, each consisting of P pixels {xi }, i = 1. [sent-55, score-0.248]

23 P , we wish to infer a segmentation S for the image. [sent-58, score-0.137]

24 S consists of a labeling si for every pixel, where si is a 1-of-(L+1) encoded variable, and L is the ﬁxed number of parts that combine to generate the foreground. [sent-59, score-0.19]

25 Accurate inference of S is driven by models for 1) part shapes and 2) part appearances. [sent-65, score-0.308]

26 Part shapes: Several types of models can be used to deﬁne probabilistic distributions over segmentations S. [sent-66, score-0.252]

27 The simplest approach is to model each pixel si independently with categorical variables whose parameters are speciﬁed by the object’s mean shape (Fig. [sent-67, score-0.365]

28 2(b)) additionally model interactions between nearby pixels using pairwise potential functions that efﬁciently capture local properties of images like smoothness and continuity. [sent-70, score-0.276]

29 2(c)) make heavy use of hidden variables to efﬁciently deﬁne higher-order potentials that take into account the conﬁguration of larger groups of image pixels. [sent-72, score-0.142]

30 In order to learn a model that accurately captures the properties of part shapes we use DBMs but also impose carefully chosen connectivity and capacity constraints, following the structure of the Shape Boltzmann Machine (SBM) [1]. [sent-76, score-0.242]

31 We further extend the model to account for multi-part shapes to obtain the Multinomial Shape Boltzmann Machine (MSBM). [sent-77, score-0.233]

32 We share weights between the four sets of ﬁrst-layer hidden units and patches, and purposely restrict the number of units in h2 . [sent-82, score-0.21]

33 These modiﬁcations signiﬁcantly reduce the number of parameters whilst taking into account an important property of shapes, namely that the strongest dependencies between pixels are typically local. [sent-83, score-0.187]

34 In all models latent units h are binary and visible units S are multinomial random variables. [sent-91, score-0.229]

35 Here we assume one background (l = 0) and two foreground (l = 1, non-body; l = 2, body) parts. [sent-96, score-0.13]

36 During pre-training, the means {µw } and covariances {⌃w } of these Gaussians are extracted by training a mixture model with W components on every pixel in the dataset, ignoring image and part structure. [sent-101, score-0.239]

37 It is also assumed that each of the L parts can have different appearances in different images, and that these appearances can be clustered into K classes. [sent-102, score-0.364]

38 Given al , the probability distribution deﬁned on pixels associated with part l is given by a Gaussian mixture model with means {µw } and covariances {⌃w } and mixing proportions { lkw }. [sent-106, score-0.343]

39 The prior on A = {al } speciﬁes the probability ⇡lk of appearance class k being chosen for part l. [sent-107, score-0.267]

40 Therefore appearance parameters ✓a = {⇡lk , lkw } (see Fig. [sent-108, score-0.315]

42 During learning, we ﬁnd the values of ✓ that maximize the likelihood of the training data D, and segmentation is performed on a previously-unseen image by querying the marginal distribution p(S|Xtest , ✓). [sent-115, score-0.229]

43 3 n H ✓a si al H xi L+1 ✓s S X A P Figure 4: A model of shape and appearance. [sent-118, score-0.345]

44 Pixels xi are modeled via appearance variables al . [sent-120, score-0.286]

45 The model’s belief about each layer’s shape is captured by shape variables H. [sent-121, score-0.416]

46 Segmentation variables si assign each pixel to a layer. [sent-122, score-0.129]

47 In order to sample p(A|S, H, X, ✓) we consider the conditional distribution of appearance class k being chosen for part l which is given by: Q P ·s ⇡lk i ( w lkw N (xi |µw , ⌃w )) li h Q P i. [sent-126, score-0.355]

48 p(alk = 1|S, X, ✓) = P (5) K ·sli r=1 ⇡lr i( w lrw N (xi |µw , ⌃w )) Since the MSBM only has edges between each pair of adjacent layers, all hidden units within a layer are conditionally independent given the units in the other two layers. [sent-127, score-0.261]

49 This property can be exploited to make inference in the shape model exact and efﬁcient. [sent-128, score-0.29]

50 The conditional probabilities are: X X 1 2 p(h1 = 1|s, h2 , ✓) = ( wlij sli + wjk h2 + c1 ), (6) j k j i,l p(h2 k 1 = 1|h , ✓) = ( X k 2 wjk h1 j + c2 ), j (7) j where (y) = 1/(1 + exp( y)) is the sigmoid function. [sent-129, score-0.371]

51 Finally, we draw samples for the pixels in p(S|A, H, X, ✓) independently: P 1 exp( j wlij h1 + bli ) p(xi |A, sli = 1, ✓) j p(sli = 1|A, H, X, ✓) = PL . [sent-132, score-0.49]

52 The computation of the likelihood p(X|✓) of image X is intractable, so we approximate the quality of each inference using a scoring function: 1X Score(X|✓) = p(X, A(t) , S(t) , H(t) |✓), (9) T t where {A(t) , S(t) , H(t) }, t = 1. [sent-135, score-0.163]

53 Learning: Learning of the model involves maximizing the log likelihood log p(D|✓a , ✓s ) of the training dataset D with respect to the model parameters ✓a and ✓s . [sent-143, score-0.128]

54 Since training is partially supervised, in that for each image X its corresponding segmentation S is also given, we can learn the parameters of the shape and appearance components separately. [sent-144, score-0.664]

55 Once this RBM is trained, we infer the conditional mean of the hidden units for each training image. [sent-150, score-0.161]

56 3 Related work Existing probabilistic models of images can be categorized by the amount of variability they expect to encounter in the data and by how they model this variability. [sent-154, score-0.241]

57 A signiﬁcant portion of the literature models images using only two parts: a foreground object and its background e. [sent-155, score-0.378]

58 Models that account for the parts within the foreground object mainly differ in how accurately they learn about and represent the variability of the shapes of the object’s parts. [sent-158, score-0.58]

59 Stel Component Analysis [10] accounts for larger amounts of shape variability by learning a number of different template means for the object that are blended together on a pixel-by-pixel basis. [sent-160, score-0.397]

60 Factored Shapes and Appearances [11] models global properties of shape using a factor analysis-like model, and ‘masked’ RBMs have been used to model more local properties of shape [20]. [sent-161, score-0.444]

61 However, none of these models constitute a strong model of shape in terms of realism of samples and generalization capabilities [1]. [sent-162, score-0.319]

62 The closest works to ours in terms of ability to deal with datasets that exhibit signiﬁcant variability in both shape and appearance are the works of Bo and Fowlkes [21] and Thomas et al. [sent-165, score-0.496]

63 Bo and Fowlkes [21] present an algorithm for pedestrian segmentation that models the shapes of the parts using several template means. [sent-167, score-0.431]

64 The different parts are composed using hand coded geometric constraints, which means that the model cannot be automatically extended to other application domains. [sent-168, score-0.118]

65 The Implicit Shape Model (ISM) used in [22] is reliant on interest point detectors and deﬁnes distributions over segmentations only in the posterior, and therefore is not fully generative. [sent-169, score-0.22]

66 Due to its modular structure, we also expect it to rapidly absorb future developments in shape and appearance models. [sent-171, score-0.435]

67 4 Experiments Penn-Fudan pedestrians: The ﬁrst dataset that we considered is Penn-Fudan pedestrians [23], consisting of 169 images of pedestrians (Fig. [sent-172, score-0.261]

68 The images are annotated with ground-truth segmentations for L = 7 different parts (hair, face, upper and lower clothes, shoes, legs, arms; Fig. [sent-174, score-0.43]

69 For the shape component, we trained an MSBM on the 684 images of a labeled version of the HumanEva dataset [24] (at 48 ⇥ 24 pixels; also ﬂipped horizontally) with overlap b = 4, and 400 and 50 hidden units in the ﬁrst and second layers respectively. [sent-177, score-0.564]

70 (a) A chain of samples (1000 samples between frames). [sent-184, score-0.115]

71 If, for example, there is a 50-50 chance that a pixel belongs to the red or blue parts, we display that pixel in purple. [sent-187, score-0.158]

72 5(b) we show the difference between the sampled shapes in Fig. [sent-197, score-0.174]

73 5(a) and their closest images in the training set (based on per-pixel label agreement). [sent-198, score-0.151]

74 We see that the model generalizes in non-trivial ways to generate realistic shapes that it had not encountered during training. [sent-199, score-0.202]

75 We used the training images in each split to train the appearance component with a vocabulary of size W = 50 and K = 100 mixture components1 . [sent-208, score-0.378]

76 We additionally constrained the model by sharing the appearance models for the arms and legs with that of the face. [sent-209, score-0.356]

77 We assess the quality of the appearance model by performing the following experiment: for each test image, we used the scoring function described in Eq. [sent-210, score-0.303]

78 9 to evaluate a number of different proposal segmentations for that image. [sent-211, score-0.22]

79 We considered 10 randomly chosen segmentations from the training dataset as well as the ground-truth segmentation for the test image, and found that the appearance model correctly assigns the highest score to the ground-truth 95% of the time. [sent-212, score-0.684]

80 During inference, the shape and appearance models (which are deﬁned on images of different sizes), were combined at 200 ⇥ 100 pixels via M ATLAB’s imresize function, and we set = 0. [sent-213, score-0.683]

81 Inference chains were seeded at 100 exemplar segmentations from the HumanEva dataset (obtained using the K-medoids algorithm with K = 100), and were run for 20 Gibbs iterations each (with 5 iterations of Eqs. [sent-216, score-0.382]

82 We compute the conditional probability of each pixel belonging to different parts given the last set of samples obtained from the highest scoring chain, assign each pixel independently to the most likely part at that pixel, and report the percentage of correctly labeled pixels (see Table 1). [sent-219, score-0.539]

83 We also report the accuracy obtained, had the top scoring seed segmentation been returned as the ﬁnal segmentation for each image. [sent-221, score-0.402]

84 Here the quality of the seed is determined solely by the appearance model. [sent-222, score-0.307]

85 We observe that the model has comparable performance to the state-of-the-art but pedestrianspeciﬁc algorithm of [21], and that inference in the model signiﬁcantly improves the accuracy of the segmentations over the baseline (top seed+SP). [sent-223, score-0.33]

86 The appearances exhibited by the parts in the dataset are highly varied, and the complexity of the appearance model reﬂects this fact. [sent-227, score-0.523]

87 We report the percentage of pixels belonging to each part that are labeled correctly. [sent-262, score-0.204]

88 8% ETHZ cars: The second dataset that we considered is the ETHZ labeled cars dataset [22], which itself is a subset of the LabelMe dataset [23], consisting of 139 images of cars, all in the same semiproﬁle view (Fig. [sent-288, score-0.28]

89 The images are annotated with ground-truth segmentations for L = 6 parts (body, wheel, window, bumper, license plate, headlight; Fig. [sent-290, score-0.507]

90 We used the training images in each split to train both the shape and appearance components. [sent-295, score-0.586]

91 For the shape component, we trained an MSBM at 50 ⇥ 50 pixels with overlap b = 4, and 2000 and 100 hidden units in the ﬁrst and second layers respectively. [sent-296, score-0.531]

92 The appearance model was trained with a vocabulary of size W = 50 and K = 100 mixture components and we set = 0. [sent-298, score-0.285]

93 Inference chains were seeded at 50 exemplar segmentations (obtained using K-medoids). [sent-300, score-0.341]

94 We ﬁnd that the use of superpixels does not help with this dataset (due to the poor quality of superpixels obtained for these images). [sent-301, score-0.133]

95 By incorporating better models of part appearance into the generative model, we expect to see this discrepancy decrease. [sent-305, score-0.337]

96 5 Conclusions and future work In this paper we have shown how the SBM can be extended to obtain the MSBM, and presented a principled probabilistic model of images of objects that exploits the MSBM as its model for part shapes. [sent-306, score-0.248]

97 We demonstrated how object segmentations can be obtained simply by performing MCMC inference in the model. [sent-307, score-0.402]

98 The model can also be treated as a probabilistic evaluator of segmentations: given a proposal segmentation it can be used to estimate its likelihood. [sent-308, score-0.197]

99 This leads us to believe that the combination of a generative model such as ours, with a discriminative, bottom-up segmentation algorithm could be highly effective. [sent-309, score-0.235]

100 We are currently investigating how textured appearance models, which take into account the spatial structure of pixels, affect the learning and inference algorithms and the performance of the model. [sent-310, score-0.312]

similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('msbm', 0.525), ('sbm', 0.241), ('appearance', 0.227), ('segmentations', 0.22), ('sli', 0.213), ('shape', 0.208), ('shapes', 0.174), ('segmentation', 0.137), ('appearances', 0.137), ('boltzmann', 0.133), ('pixels', 0.128), ('object', 0.128), ('fowlkes', 0.127), ('images', 0.12), ('ism', 0.107), ('foreground', 0.096), ('parts', 0.09), ('lkw', 0.088), ('nicolas', 0.085), ('units', 0.08), ('seed', 0.08), ('pixel', 0.079), ('bo', 0.078), ('ethz', 0.077), ('license', 0.077), ('vittorio', 0.071), ('generative', 0.07), ('alk', 0.066), ('bumper', 0.066), ('eslami', 0.066), ('nebojsa', 0.066), ('wlij', 0.066), ('body', 0.063), ('image', 0.061), ('rbms', 0.061), ('variability', 0.061), ('al', 0.059), ('charless', 0.058), ('exemplar', 0.058), ('wheel', 0.058), ('inference', 0.054), ('arms', 0.053), ('layer', 0.051), ('lk', 0.051), ('hidden', 0.05), ('cvpr', 0.05), ('dbms', 0.05), ('atlab', 0.05), ('heess', 0.05), ('pedestrians', 0.05), ('john', 0.05), ('si', 0.05), ('thomas', 0.05), ('winn', 0.048), ('scoring', 0.048), ('legs', 0.048), ('sp', 0.047), ('superpixels', 0.046), ('wjk', 0.046), ('bli', 0.044), ('humaneva', 0.044), ('realism', 0.044), ('shoes', 0.044), ('stel', 0.044), ('ali', 0.043), ('dataset', 0.041), ('part', 0.04), ('multinomial', 0.04), ('rbm', 0.039), ('edinburgh', 0.039), ('samples', 0.039), ('hair', 0.039), ('headlight', 0.039), ('jamie', 0.039), ('cars', 0.037), ('chain', 0.037), ('christopher', 0.036), ('belonging', 0.036), ('jojic', 0.036), ('layers', 0.035), ('background', 0.034), ('masked', 0.033), ('probabilistic', 0.032), ('seeded', 0.032), ('training', 0.031), ('vision', 0.031), ('deep', 0.031), ('account', 0.031), ('chains', 0.031), ('bg', 0.03), ('labelme', 0.03), ('luc', 0.03), ('pedestrian', 0.03), ('trained', 0.03), ('mcmc', 0.029), ('visible', 0.029), ('model', 0.028), ('ferrari', 0.028), ('interactive', 0.028), ('whilst', 0.028)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 1.0 8 nips-2012-A Generative Model for Parts-based Object Segmentation

Author: S. Eslami, Christopher Williams

2 0.17047296 40 nips-2012-Analyzing 3D Objects in Cluttered Images

Author: Mohsen Hejrati, Deva Ramanan

Abstract: We present an approach to detecting and analyzing the 3D conﬁguration of objects in real-world images with heavy occlusion and clutter. We focus on the application of ﬁnding and analyzing cars. We do so with a two-stage model; the ﬁrst stage reasons about 2D shape and appearance variation due to within-class variation (station wagons look different than sedans) and changes in viewpoint. Rather than using a view-based model, we describe a compositional representation that models a large number of effective views and shapes using a small number of local view-based templates. We use this model to propose candidate detections and 2D estimates of shape. These estimates are then reﬁned by our second stage, using an explicit 3D model of shape and viewpoint. We use a morphable model to capture 3D within-class variation, and use a weak-perspective camera model to capture viewpoint. We learn all model parameters from 2D annotations. We demonstrate state-of-the-art accuracy for detection, viewpoint estimation, and 3D shape reconstruction on challenging images from the PASCAL VOC 2011 dataset. 1

3 0.14170952 91 nips-2012-Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images

Author: Dan Ciresan, Alessandro Giusti, Luca M. Gambardella, Jürgen Schmidhuber

Abstract: We address a central problem of neuroanatomy, namely, the automatic segmentation of neuronal structures depicted in stacks of electron microscopy (EM) images. This is necessary to efﬁciently map 3D brain structure and connectivity. To segment biological neuron membranes, we use a special type of deep artiﬁcial neural network as a pixel classiﬁer. The label of each pixel (membrane or nonmembrane) is predicted from raw pixel values in a square window centered on it. The input layer maps each window pixel to a neuron. It is followed by a succession of convolutional and max-pooling layers which preserve 2D information and extract features with increasing levels of abstraction. The output layer produces a calibrated probability for each class. The classiﬁer is trained by plain gradient descent on a 512 × 512 × 30 stack with known ground truth, and tested on a stack of the same size (ground truth unknown to the authors) by the organizers of the ISBI 2012 EM Segmentation Challenge. Even without problem-speciﬁc postprocessing, our approach outperforms competing techniques by a large margin in all three considered metrics, i.e. rand error, warping error and pixel error. For pixel error, our approach is the only one outperforming a second human observer. 1

4 0.1156229 4 nips-2012-A Better Way to Pretrain Deep Boltzmann Machines

Author: Geoffrey E. Hinton, Ruslan Salakhutdinov

Abstract: We describe how the pretraining algorithm for Deep Boltzmann Machines (DBMs) is related to the pretraining algorithm for Deep Belief Networks and we show that under certain conditions, the pretraining procedure improves the variational lower bound of a two-hidden-layer DBM. Based on this analysis, we develop a different method of pretraining DBMs that distributes the modelling work more evenly over the hidden layers. Our results on the MNIST and NORB datasets demonstrate that the new pretraining algorithm allows us to learn better generative models. 1

5 0.10828745 357 nips-2012-Unsupervised Template Learning for Fine-Grained Object Recognition

Author: Shulin Yang, Liefeng Bo, Jue Wang, Linda G. Shapiro

Abstract: Fine-grained recognition refers to a subordinate level of recognition, such as recognizing different species of animals and plants. It differs from recognition of basic categories, such as humans, tables, and computers, in that there are global similarities in shape and structure shared cross different categories, and the differences are in the details of object parts. We suggest that the key to identifying the ﬁne-grained differences lies in ﬁnding the right alignment of image regions that contain the same object parts. We propose a template model for the purpose, which captures common shape patterns of object parts, as well as the cooccurrence relation of the shape patterns. Once the image regions are aligned, extracted features are used for classiﬁcation. Learning of the template model is efﬁcient, and the recognition results we achieve signiﬁcantly outperform the stateof-the-art algorithms. 1

6 0.10795651 229 nips-2012-Multimodal Learning with Deep Boltzmann Machines

7 0.10493867 106 nips-2012-Dynamical And-Or Graph Learning for Object Shape Modeling and Detection

8 0.10215084 137 nips-2012-From Deformations to Parts: Motion-based Segmentation of 3D Objects

9 0.097764857 158 nips-2012-ImageNet Classification with Deep Convolutional Neural Networks

10 0.096493483 201 nips-2012-Localizing 3D cuboids in single-view images

11 0.096343338 65 nips-2012-Cardinality Restricted Boltzmann Machines

12 0.094322897 303 nips-2012-Searching for objects driven by context

13 0.093508504 81 nips-2012-Context-Sensitive Decision Forests for Object Detection

14 0.092448995 1 nips-2012-3D Object Detection and Viewpoint Estimation with a Deformable 3D Cuboid Model

15 0.089117043 344 nips-2012-Timely Object Recognition

16 0.082102485 92 nips-2012-Deep Representations and Codes for Image Auto-Annotation

17 0.080648996 230 nips-2012-Multiple Choice Learning: Learning to Produce Multiple Structured Outputs

18 0.08024434 87 nips-2012-Convolutional-Recursive Deep Learning for 3D Object Classification

19 0.079767615 193 nips-2012-Learning to Align from Scratch

20 0.078567356 82 nips-2012-Continuous Relaxations for Discrete Hamiltonian Monte Carlo

similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.165), (1, 0.039), (2, -0.212), (3, 0.002), (4, 0.062), (5, -0.089), (6, -0.002), (7, -0.096), (8, 0.005), (9, -0.021), (10, -0.017), (11, 0.063), (12, 0.029), (13, -0.038), (14, -0.027), (15, 0.058), (16, 0.013), (17, -0.115), (18, 0.019), (19, -0.102), (20, 0.048), (21, -0.097), (22, -0.019), (23, 0.112), (24, -0.011), (25, 0.043), (26, 0.044), (27, -0.011), (28, 0.029), (29, -0.032), (30, -0.015), (31, -0.079), (32, -0.099), (33, 0.007), (34, -0.002), (35, 0.0), (36, -0.054), (37, -0.009), (38, 0.018), (39, -0.033), (40, -0.042), (41, -0.043), (42, -0.035), (43, -0.079), (44, 0.041), (45, 0.017), (46, -0.022), (47, -0.027), (48, 0.098), (49, -0.113)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.94565701 8 nips-2012-A Generative Model for Parts-based Object Segmentation

Author: S. Eslami, Christopher Williams

2 0.74800026 40 nips-2012-Analyzing 3D Objects in Cluttered Images

Author: Mohsen Hejrati, Deva Ramanan

3 0.72009689 357 nips-2012-Unsupervised Template Learning for Fine-Grained Object Recognition

Author: Shulin Yang, Liefeng Bo, Jue Wang, Linda G. Shapiro

4 0.70547414 201 nips-2012-Localizing 3D cuboids in single-view images

Author: Jianxiong Xiao, Bryan Russell, Antonio Torralba

Abstract: In this paper we seek to detect rectangular cuboids and localize their corners in uncalibrated single-view images depicting everyday scenes. In contrast to recent approaches that rely on detecting vanishing points of the scene and grouping line segments to form cuboids, we build a discriminative parts-based detector that models the appearance of the cuboid corners and internal edges while enforcing consistency to a 3D cuboid model. Our model copes with different 3D viewpoints and aspect ratios and is able to detect cuboids across many different object categories. We introduce a database of images with cuboid annotations that spans a variety of indoor and outdoor scenes and show qualitative and quantitative results on our collected database. Our model out-performs baseline detectors that use 2D constraints alone on the task of localizing cuboid corners. 1

5 0.66232121 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

6 0.64501852 1 nips-2012-3D Object Detection and Viewpoint Estimation with a Deformable 3D Cuboid Model

7 0.63653958 235 nips-2012-Natural Images, Gaussian Mixtures and Dead Leaves

8 0.63167441 137 nips-2012-From Deformations to Parts: Motion-based Segmentation of 3D Objects

9 0.62396795 91 nips-2012-Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images

10 0.62235779 210 nips-2012-Memorability of Image Regions

11 0.6045633 65 nips-2012-Cardinality Restricted Boltzmann Machines

12 0.59372127 158 nips-2012-ImageNet Classification with Deep Convolutional Neural Networks

13 0.59177715 159 nips-2012-Image Denoising and Inpainting with Deep Neural Networks

14 0.59092528 229 nips-2012-Multimodal Learning with Deep Boltzmann Machines

15 0.57212603 185 nips-2012-Learning about Canonical Views from Internet Image Collections

16 0.5683676 87 nips-2012-Convolutional-Recursive Deep Learning for 3D Object Classification

17 0.5492056 146 nips-2012-Graphical Gaussian Vector for Image Categorization

18 0.54850602 303 nips-2012-Searching for objects driven by context

19 0.54092741 106 nips-2012-Dynamical And-Or Graph Learning for Object Shape Modeling and Detection

20 0.52963406 101 nips-2012-Discriminatively Trained Sparse Code Gradients for Contour Detection

similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(0, 0.023), (2, 0.254), (17, 0.011), (21, 0.035), (36, 0.011), (38, 0.073), (39, 0.012), (42, 0.024), (44, 0.017), (53, 0.011), (54, 0.025), (55, 0.022), (74, 0.147), (76, 0.103), (80, 0.088), (92, 0.065)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.77424043 8 nips-2012-A Generative Model for Parts-based Object Segmentation

Author: S. Eslami, Christopher Williams

2 0.69909537 196 nips-2012-Learning with Partially Absorbing Random Walks

Author: Xiao-ming Wu, Zhenguo Li, Anthony M. So, John Wright, Shih-fu Chang

Abstract: We propose a novel stochastic process that is with probability αi being absorbed at current state i, and with probability 1 − αi follows a random edge out of it. We analyze its properties and show its potential for exploring graph structures. We prove that under proper absorption rates, a random walk starting from a set S of low conductance will be mostly absorbed in S. Moreover, the absorption probabilities vary slowly inside S, while dropping sharply outside, thus implementing the desirable cluster assumption for graph-based learning. Remarkably, the partially absorbing process uniﬁes many popular models arising in a variety of contexts, provides new insights into them, and makes it possible for transferring ﬁndings from one paradigm to another. Simulation results demonstrate its promising applications in retrieval and classiﬁcation.

3 0.64388317 37 nips-2012-Affine Independent Variational Inference

Author: Edward Challis, David Barber

Abstract: We consider inference in a broad class of non-conjugate probabilistic models based on minimising the Kullback-Leibler divergence between the given target density and an approximating ‘variational’ density. In particular, for generalised linear models we describe approximating densities formed from an afﬁne transformation of independently distributed latent variables, this class including many well known densities as special cases. We show how all relevant quantities can be efﬁciently computed using the fast Fourier transform. This extends the known class of tractable variational approximations and enables the ﬁtting for example of skew variational densities to the target density. 1

4 0.63596165 3 nips-2012-A Bayesian Approach for Policy Learning from Trajectory Preference Queries

Author: Aaron Wilson, Alan Fern, Prasad Tadepalli

Abstract: We consider the problem of learning control policies via trajectory preference queries to an expert. In particular, the agent presents an expert with short runs of a pair of policies originating from the same state and the expert indicates which trajectory is preferred. The agent’s goal is to elicit a latent target policy from the expert with as few queries as possible. To tackle this problem we propose a novel Bayesian model of the querying process and introduce two methods that exploit this model to actively select expert queries. Experimental results on four benchmark problems indicate that our model can effectively learn policies from trajectory preference queries and that active query selection can be substantially more efﬁcient than random selection. 1

5 0.6332258 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 0.63236231 202 nips-2012-Locally Uniform Comparison Image Descriptor

7 0.62881875 360 nips-2012-Visual Recognition using Embedded Feature Selection for Curvature Self-Similarity

8 0.62599301 274 nips-2012-Priors for Diversity in Generative Latent Variable Models

9 0.62458426 40 nips-2012-Analyzing 3D Objects in Cluttered Images

10 0.61662328 337 nips-2012-The Lovász ϑ function, SVMs and finding large dense subgraphs

11 0.61624265 321 nips-2012-Spectral learning of linear dynamics from generalised-linear observations with application to neural population data

12 0.61367708 17 nips-2012-A Scalable CUR Matrix Decomposition Algorithm: Lower Time Complexity and Tighter Bound

13 0.61163723 101 nips-2012-Discriminatively Trained Sparse Code Gradients for Contour Detection

14 0.61110801 201 nips-2012-Localizing 3D cuboids in single-view images

15 0.60977763 176 nips-2012-Learning Image Descriptors with the Boosting-Trick

16 0.60758543 357 nips-2012-Unsupervised Template Learning for Fine-Grained Object Recognition

17 0.60497087 168 nips-2012-Kernel Latent SVM for Visual Recognition

18 0.60436803 210 nips-2012-Memorability of Image Regions

19 0.6036247 193 nips-2012-Learning to Align from Scratch

20 0.5937742 235 nips-2012-Natural Images, Gaussian Mixtures and Dead Leaves