iccv iccv2013 iccv2013-162 knowledge-graph by maker-knowledge-mining

162 iccv-2013-Fast Subspace Search via Grassmannian Based Hashing


Source: pdf

Author: Xu Wang, Stefan Atev, John Wright, Gilad Lerman

Abstract: The problem of efficiently deciding which of a database of models is most similar to a given input query arises throughout modern computer vision. Motivated by applications in recognition, image retrieval and optimization, there has been significant recent interest in the variant of this problem in which the database models are linear subspaces and the input is either a point or a subspace. Current approaches to this problem have poor scaling in high dimensions, and may not guarantee sublinear query complexity. We present a new approach to approximate nearest subspace search, based on a simple, new locality sensitive hash for subspaces. Our approach allows point-tosubspace query for a database of subspaces of arbitrary dimension d, in a time that depends sublinearly on the number of subspaces in the database. The query complexity of our algorithm is linear in the ambient dimension D, allow- ing it to be directly applied to high-dimensional imagery data. Numerical experiments on model problems in image repatching and automatic face recognition confirm the advantages of our algorithm in terms of both speed and accuracy.

Reference: text


Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 edu John Wright EE Department, Columbia University j ohnwri ght @ ee . [sent-3, score-0.079]

2 edu Abstract The problem of efficiently deciding which of a database of models is most similar to a given input query arises throughout modern computer vision. [sent-5, score-0.738]

3 Motivated by applications in recognition, image retrieval and optimization, there has been significant recent interest in the variant of this problem in which the database models are linear subspaces and the input is either a point or a subspace. [sent-6, score-0.445]

4 Current approaches to this problem have poor scaling in high dimensions, and may not guarantee sublinear query complexity. [sent-7, score-0.777]

5 We present a new approach to approximate nearest subspace search, based on a simple, new locality sensitive hash for subspaces. [sent-8, score-0.695]

6 Our approach allows point-tosubspace query for a database of subspaces of arbitrary dimension d, in a time that depends sublinearly on the number of subspaces in the database. [sent-9, score-1.419]

7 The query complexity of our algorithm is linear in the ambient dimension D, allow- ing it to be directly applied to high-dimensional imagery data. [sent-10, score-0.825]

8 Numerical experiments on model problems in image repatching and automatic face recognition confirm the advantages of our algorithm in terms of both speed and accuracy. [sent-11, score-0.17]

9 Introduction Given a very large database of models, how can we efficiently determine which one that best fits a given input query? [sent-13, score-0.113]

10 This basic question arises repeatedly in computer vision applications such as visual recognition, categorization, image retrieval and beyond. [sent-14, score-0.046]

11 These applications pose two general challenges to the algorithm designer: imagery data (and their features) are typically high-dimensional, and databases arising in applications can be very large scale. [sent-15, score-0.093]

12 The large scale often precludes simply comparing the query to each of the models in any reasonable amount of time. [sent-16, score-0.581]

13 edu ticated approximate nearest neighbor techniques, whose query time is sublinear in the size of the database. [sent-20, score-1.061]

14 For the case in which the query is a vector and the database is also a collection of vectors, these techniques are very well- developed, in both theory and practice [5, 6, 11, 21]. [sent-21, score-0.722]

15 However, data in computer vision problems often have rich physical or geometric structure, which may not be wellencoded using point models. [sent-22, score-0.039]

16 For example, photometric or textural properties of a collection of images can often be better represented using linear or affine subspaces, rather than a simple point model. [sent-23, score-0.148]

17 In the approximate nearest subspace problem, we are given a collection of linear subspaces. [sent-24, score-0.527]

18 The goal is to efficiently determine which of the database subspaces is closest to the input [3]. [sent-25, score-0.406]

19 Good solutions to this problem would allow us to efficiently query large databases which contain much richer representations. [sent-26, score-0.576]

20 In contrast to approximate nearest neighbor, both the theory and practice of approximate nearest subspace are still developing. [sent-27, score-0.754]

21 It maps each subspace S ⊆ RD to its orthogonal projection mmaaptrsix e aPchS, aunbds pthacene applies an approximate nearest neighbor algorithm to the projection matrices. [sent-31, score-0.542]

22 The advantage of this approach is that it cleanly reduces the subspace problem to the better-understood point search problem. [sent-32, score-0.428]

23 1 Moreover, the mapping from a subspace to an orthoprojector does not preserve distances (for subspaces of different dimensions), and so performance guarantees for the approximate nearest neighbor algorithm may not pull back to the approximate nearest subspace problem. [sent-34, score-1.4]

24 Algorithms with sublinear query time, but exponential dependence on dimension have also been introduced in [16]. [sent-35, score-0.963]

25 Motivated in part by [3], there has been a flurry of recent work on special cases of this problem. [sent-36, score-0.086]

26 For example the 1[3] suggest using random projections after lifting trolling the complexity. [sent-37, score-0.039]

27 as one means of con- 2776 case in which the query is a point and the database contains hyperplanes (of dimension d = D 1) has been studied ihny pcoernpnlaencteiosn ( tof daicmtivenes learning a Dnd − large-scale regression [13]. [sent-38, score-0.897]

28 Various approaches based on locality sensitive hashing have been proposed [13, 15, 18, 19]. [sent-39, score-0.364]

29 While various technical obstacles prevent these approaches from guaranteeing sublinear query complexity over all inputs, they have been used effectively in various practical vision problems. [sent-40, score-0.861]

30 In the algorithms community, there is also dedicated work on the special case in which the queries are points and the database consists of affine lines (d = 1). [sent-41, score-0.227]

31 produce a data structure for this problem that has query time O(D3n1/2+t) and space complexity for any t > 0 [2]. [sent-44, score-0.537]

32 Again, the query time O(D3) could be problematic in large dimensions. [sent-45, score-0.537]

33 Moreover, many of the most interesting models for computer vision have a dimension d that falls somewhere in between 1and D 1. [sent-46, score-0.17]

34 For example, linear subspaces spanned by images dta Dken − u1. [sent-47, score-0.337]

35 Local image patches also typically lie near subspaces of dimension higher than one [22, 23]. [sent-49, score-0.426]

36 In this paper, we provide a solution of the approximate nearest subspace (ANS) search problem based on the notion of locality sensitive hashing (see e. [sent-52, score-0.938]

37 -Fsourb asplla cofe tqhueesrey s (efaorrc shuebs-, our preprocessing space is O(n1+ρ + nDd) and query time is O(Ddnρ), where d is the largest dimension of subspaces among both query elements and the database elements, D is the ambient dimension and ρ < 1. [sent-57, score-1.94]

38 For line-line search we can obtain better estimates of the parameters, in particular, asymptotic estimate of ρ (for a special setting). [sent-60, score-0.161]

39 Our theoretical setting is designed to address recognition problems. [sent-61, score-0.049]

40 For example, our unorthodox restriction on the maximal distance between query element and the database (this appears only in some of our statements) can often be met in practice, where query points may be contained in or be sufficiently close to the database. [sent-62, score-1.355]

41 We confirmed in practice the competitive speed and accuracy ofour proposed solution on model problems in image repatching and automatic face recognition. [sent-63, score-0.204]

42 In §2 we introduce notational conventions and adapt the Innot i§o2n wofe locality sceen nsitoitvaehashing to the ANS problem. [sent-65, score-0.243]

43 We then generalize a wellknown theoretical framework claiming that a locality sensitive hashing family gives rise to a search algorithm with sub-linear time. [sent-66, score-0.591]

44 In §3, we propose a concrete hashing family f-olirn tehaer G tirmaes. [sent-67, score-0.266]

45 We then formulate the main theorems Gof( tDhi,s1 )w ∪or Gk detailing eth teh quality uolaf teh teh eb masiaci s tuhbe-olirneemars search procedure in each one of the three types of searches described above. [sent-69, score-0.304]

46 The details of the ANS algorithm resulting from the locality sensitive hashing family we proposed are outlines in §4, whereas §5 compares our ANS algorithm mareeth ouodtl nweisth i nth §e4 A, wNhSe algorithm mofp Baraessri o uetr a Al. [sent-70, score-0.503]

47 N[3S] on omriothdmel problems in image repatching and automatic face recognition. [sent-71, score-0.17]

48 , the space of all d-dimensional linear subspaces of RD. [sent-76, score-0.293]

49 If 0 < d1 ≤ d2 < D, L1 ∈ G(D, d1) and L2 ∈ G(D, d2), the princi≤pal d angles θ1 ≥ . [sent-77, score-0.057]

50 For example, =if dd1 = 1, then distG (L1, L2) is the elevation angle between the line L1 and the subspace L2. [sent-97, score-0.34]

51 2Here, order the principal angles decreasingly, unlike the [9] (§ 12. [sent-98, score-0.103]

52 Motivated by the approximate nearest point search in [1], we define the approximate nearest subspace search problem as follows: Definition 2. [sent-102, score-0.929]

53 (R, c)-approximate subspace search: Let X be a set of d2-dimensional subspaces in RD and R, c, δ be positive numbers. [sent-104, score-0.551]

54 A search algorithm is called (R, c)approximate subspace search if it fulfills the following requirement. [sent-105, score-0.465]

55 Given a query subspace L of dimension d1, if there is an element L? [sent-106, score-0.995]

56 − For several applications, the most interesting query problem is the point-subspace query, the query is a point in RD and the database is a subset of G(D, d). [sent-115, score-1.226]

57 By connecting points with the origin to obtain lines, the point-subspace query problem is reduced to the (R, c)-approximate subspace search problem with d1 = 1 (where we denote d2 by d). [sent-116, score-0.88]

58 However, instead of measuring the Euclidean distance of the query point to the subspace, we measure the equivalent “distance”, distG, between the line through the query point and the subspace. [sent-117, score-1.207]

59 The use of this equivalent “distance” results in a pointsubspace query. [sent-118, score-0.055]

60 Indeed, assume that the query point x0 has a principal angle θ0 and Euclidean distance | |x0 | |2 sin θ0 w. [sent-119, score-0.732]


similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('query', 0.537), ('subspaces', 0.293), ('subspace', 0.258), ('sublinear', 0.24), ('distg', 0.232), ('repatching', 0.17), ('hashing', 0.158), ('locality', 0.149), ('ans', 0.134), ('dimension', 0.133), ('nearest', 0.128), ('database', 0.113), ('minnesota', 0.113), ('approximate', 0.103), ('ambient', 0.101), ('umn', 0.1), ('lerman', 0.1), ('search', 0.085), ('math', 0.069), ('sin', 0.068), ('element', 0.067), ('family', 0.058), ('angles', 0.057), ('sensitive', 0.057), ('equivalent', 0.055), ('imagery', 0.054), ('neighbor', 0.053), ('dependence', 0.053), ('sufficiently', 0.051), ('searches', 0.05), ('unorthodox', 0.05), ('ndc', 0.05), ('dnd', 0.05), ('cofe', 0.05), ('sublinearly', 0.05), ('animde', 0.05), ('conventions', 0.05), ('ddn', 0.05), ('dken', 0.05), ('flurry', 0.05), ('ishe', 0.05), ('ndd', 0.05), ('rdix', 0.05), ('tehaer', 0.05), ('theoretical', 0.049), ('department', 0.047), ('gof', 0.046), ('aoinf', 0.046), ('tli', 0.046), ('agn', 0.046), ('cleanly', 0.046), ('mofp', 0.046), ('proto', 0.046), ('rd', 0.046), ('arises', 0.046), ('motivated', 0.046), ('principal', 0.046), ('teh', 0.044), ('queries', 0.044), ('wofe', 0.044), ('dta', 0.044), ('precludes', 0.044), ('designer', 0.044), ('ght', 0.044), ('preprocessing', 0.043), ('guarantees', 0.042), ('angle', 0.042), ('euclidean', 0.042), ('deciding', 0.042), ('guaranteeing', 0.042), ('ret', 0.042), ('obstacles', 0.042), ('elevation', 0.04), ('tohe', 0.04), ('pal', 0.04), ('asymptotic', 0.04), ('andoni', 0.04), ('stefan', 0.04), ('point', 0.039), ('databases', 0.039), ('tof', 0.039), ('grassmannian', 0.039), ('lifting', 0.039), ('collection', 0.038), ('theorems', 0.037), ('fulfills', 0.037), ('somewhere', 0.037), ('arccos', 0.037), ('textural', 0.037), ('special', 0.036), ('hyperplanes', 0.036), ('tehde', 0.036), ('outlines', 0.035), ('statements', 0.035), ('wellknown', 0.035), ('ee', 0.035), ('pull', 0.034), ('tthhe', 0.034), ('practice', 0.034), ('affine', 0.034)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.9999997 162 iccv-2013-Fast Subspace Search via Grassmannian Based Hashing

Author: Xu Wang, Stefan Atev, John Wright, Gilad Lerman

Abstract: The problem of efficiently deciding which of a database of models is most similar to a given input query arises throughout modern computer vision. Motivated by applications in recognition, image retrieval and optimization, there has been significant recent interest in the variant of this problem in which the database models are linear subspaces and the input is either a point or a subspace. Current approaches to this problem have poor scaling in high dimensions, and may not guarantee sublinear query complexity. We present a new approach to approximate nearest subspace search, based on a simple, new locality sensitive hash for subspaces. Our approach allows point-tosubspace query for a database of subspaces of arbitrary dimension d, in a time that depends sublinearly on the number of subspaces in the database. The query complexity of our algorithm is linear in the ambient dimension D, allow- ing it to be directly applied to high-dimensional imagery data. Numerical experiments on model problems in image repatching and automatic face recognition confirm the advantages of our algorithm in terms of both speed and accuracy.

2 0.37478748 266 iccv-2013-Mining Multiple Queries for Image Retrieval: On-the-Fly Learning of an Object-Specific Mid-level Representation

Author: Basura Fernando, Tinne Tuytelaars

Abstract: In this paper we present a new method for object retrieval starting from multiple query images. The use of multiple queries allows for a more expressive formulation of the query object including, e.g., different viewpoints and/or viewing conditions. This, in turn, leads to more diverse and more accurate retrieval results. When no query images are available to the user, they can easily be retrieved from the internet using a standard image search engine. In particular, we propose a new method based on pattern mining. Using the minimal description length principle, we derive the most suitable set of patterns to describe the query object, with patterns corresponding to local feature configurations. This results in apowerful object-specific mid-level image representation. The archive can then be searched efficiently for similar images based on this representation, using a combination of two inverted file systems. Since the patterns already encode local spatial information, good results on several standard image retrieval datasets are obtained even without costly re-ranking based on geometric verification.

3 0.35702196 337 iccv-2013-Random Grids: Fast Approximate Nearest Neighbors and Range Searching for Image Search

Author: Dror Aiger, Efi Kokiopoulou, Ehud Rivlin

Abstract: We propose two solutions for both nearest neighbors and range search problems. For the nearest neighbors problem, we propose a c-approximate solutionfor the restricted version ofthe decisionproblem with bounded radius which is then reduced to the nearest neighbors by a known reduction. For range searching we propose a scheme that learns the parameters in a learning stage adopting them to the case of a set of points with low intrinsic dimension that are embedded in high dimensional space (common scenario for image point descriptors). We compare our algorithms to the best known methods for these problems, i.e. LSH, ANN and FLANN. We show analytically and experimentally that we can do better for moderate approximation factor. Our algorithms are trivial to parallelize. In the experiments conducted, running on couple of million im- ages, our algorithms show meaningful speed-ups when compared with the above mentioned methods.

4 0.23287445 360 iccv-2013-Robust Subspace Clustering via Half-Quadratic Minimization

Author: Yingya Zhang, Zhenan Sun, Ran He, Tieniu Tan

Abstract: Subspace clustering has important and wide applications in computer vision and pattern recognition. It is a challenging task to learn low-dimensional subspace structures due to the possible errors (e.g., noise and corruptions) existing in high-dimensional data. Recent subspace clustering methods usually assume a sparse representation of corrupted errors and correct the errors iteratively. However large corruptions in real-world applications can not be well addressed by these methods. A novel optimization model for robust subspace clustering is proposed in this paper. The objective function of our model mainly includes two parts. The first part aims to achieve a sparse representation of each high-dimensional data point with other data points. The second part aims to maximize the correntropy between a given data point and its low-dimensional representation with other points. Correntropy is a robust measure so that the influence of large corruptions on subspace clustering can be greatly suppressed. An extension of our method with explicit introduction of representation error terms into the model is also proposed. Half-quadratic minimization is provided as an efficient solution to the proposed robust subspace clustering formulations. Experimental results on Hopkins 155 dataset and Extended Yale Database B demonstrate that our method outperforms state-of-the-art subspace clustering methods.

5 0.21455002 334 iccv-2013-Query-Adaptive Asymmetrical Dissimilarities for Visual Object Retrieval

Author: Cai-Zhi Zhu, Hervé Jégou, Shin'Ichi Satoh

Abstract: Visual object retrieval aims at retrieving, from a collection of images, all those in which a given query object appears. It is inherently asymmetric: the query object is mostly included in the database image, while the converse is not necessarily true. However, existing approaches mostly compare the images with symmetrical measures, without considering the different roles of query and database. This paper first measure the extent of asymmetry on large-scale public datasets reflecting this task. Considering the standard bag-of-words representation, we then propose new asymmetrical dissimilarities accounting for the different inlier ratios associated with query and database images. These asymmetrical measures depend on the query, yet they are compatible with an inverted file structure, without noticeably impacting search efficiency. Our experiments show the benefit of our approach, and show that the visual object retrieval task is better treated asymmetrically, in the spirit of state-of-the-art text retrieval.

6 0.19533756 450 iccv-2013-What is the Most EfficientWay to Select Nearest Neighbor Candidates for Fast Approximate Nearest Neighbor Search?

7 0.19338256 400 iccv-2013-Stable Hyper-pooling and Query Expansion for Event Detection

8 0.18325683 232 iccv-2013-Latent Space Sparse Subspace Clustering

9 0.18029316 444 iccv-2013-Viewing Real-World Faces in 3D

10 0.17002976 233 iccv-2013-Latent Task Adaptation with Large-Scale Hierarchies

11 0.16594146 229 iccv-2013-Large-Scale Video Hashing via Structure Learning

12 0.16341215 210 iccv-2013-Image Retrieval Using Textual Cues

13 0.16123009 83 iccv-2013-Complementary Projection Hashing

14 0.14677811 333 iccv-2013-Quantize and Conquer: A Dimensionality-Recursive Solution to Clustering, Vector Quantization, and Image Retrieval

15 0.14393097 378 iccv-2013-Semantic-Aware Co-indexing for Image Retrieval

16 0.1335385 438 iccv-2013-Unsupervised Visual Domain Adaptation Using Subspace Alignment

17 0.13164577 159 iccv-2013-Fast Neighborhood Graph Search Using Cartesian Concatenation

18 0.11915601 239 iccv-2013-Learning Hash Codes with Listwise Supervision

19 0.11901881 235 iccv-2013-Learning Coupled Feature Spaces for Cross-Modal Matching

20 0.11626311 182 iccv-2013-GOSUS: Grassmannian Online Subspace Updates with Structured-Sparsity


similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.156), (1, 0.048), (2, -0.128), (3, -0.138), (4, -0.095), (5, 0.359), (6, 0.082), (7, 0.043), (8, -0.103), (9, 0.164), (10, 0.147), (11, 0.028), (12, -0.022), (13, 0.062), (14, -0.036), (15, -0.06), (16, 0.108), (17, -0.198), (18, 0.144), (19, -0.087), (20, -0.042), (21, -0.011), (22, -0.021), (23, -0.098), (24, -0.076), (25, -0.054), (26, -0.033), (27, -0.08), (28, 0.001), (29, -0.051), (30, -0.07), (31, 0.03), (32, 0.025), (33, -0.075), (34, 0.036), (35, -0.025), (36, -0.044), (37, 0.041), (38, -0.021), (39, -0.023), (40, 0.061), (41, 0.006), (42, 0.044), (43, 0.01), (44, 0.033), (45, 0.03), (46, -0.004), (47, 0.062), (48, 0.002), (49, -0.044)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.98679113 162 iccv-2013-Fast Subspace Search via Grassmannian Based Hashing

Author: Xu Wang, Stefan Atev, John Wright, Gilad Lerman

Abstract: The problem of efficiently deciding which of a database of models is most similar to a given input query arises throughout modern computer vision. Motivated by applications in recognition, image retrieval and optimization, there has been significant recent interest in the variant of this problem in which the database models are linear subspaces and the input is either a point or a subspace. Current approaches to this problem have poor scaling in high dimensions, and may not guarantee sublinear query complexity. We present a new approach to approximate nearest subspace search, based on a simple, new locality sensitive hash for subspaces. Our approach allows point-tosubspace query for a database of subspaces of arbitrary dimension d, in a time that depends sublinearly on the number of subspaces in the database. The query complexity of our algorithm is linear in the ambient dimension D, allow- ing it to be directly applied to high-dimensional imagery data. Numerical experiments on model problems in image repatching and automatic face recognition confirm the advantages of our algorithm in terms of both speed and accuracy.

2 0.85492069 337 iccv-2013-Random Grids: Fast Approximate Nearest Neighbors and Range Searching for Image Search

Author: Dror Aiger, Efi Kokiopoulou, Ehud Rivlin

Abstract: We propose two solutions for both nearest neighbors and range search problems. For the nearest neighbors problem, we propose a c-approximate solutionfor the restricted version ofthe decisionproblem with bounded radius which is then reduced to the nearest neighbors by a known reduction. For range searching we propose a scheme that learns the parameters in a learning stage adopting them to the case of a set of points with low intrinsic dimension that are embedded in high dimensional space (common scenario for image point descriptors). We compare our algorithms to the best known methods for these problems, i.e. LSH, ANN and FLANN. We show analytically and experimentally that we can do better for moderate approximation factor. Our algorithms are trivial to parallelize. In the experiments conducted, running on couple of million im- ages, our algorithms show meaningful speed-ups when compared with the above mentioned methods.

3 0.85378855 266 iccv-2013-Mining Multiple Queries for Image Retrieval: On-the-Fly Learning of an Object-Specific Mid-level Representation

Author: Basura Fernando, Tinne Tuytelaars

Abstract: In this paper we present a new method for object retrieval starting from multiple query images. The use of multiple queries allows for a more expressive formulation of the query object including, e.g., different viewpoints and/or viewing conditions. This, in turn, leads to more diverse and more accurate retrieval results. When no query images are available to the user, they can easily be retrieved from the internet using a standard image search engine. In particular, we propose a new method based on pattern mining. Using the minimal description length principle, we derive the most suitable set of patterns to describe the query object, with patterns corresponding to local feature configurations. This results in apowerful object-specific mid-level image representation. The archive can then be searched efficiently for similar images based on this representation, using a combination of two inverted file systems. Since the patterns already encode local spatial information, good results on several standard image retrieval datasets are obtained even without costly re-ranking based on geometric verification.

4 0.79575914 334 iccv-2013-Query-Adaptive Asymmetrical Dissimilarities for Visual Object Retrieval

Author: Cai-Zhi Zhu, Hervé Jégou, Shin'Ichi Satoh

Abstract: Visual object retrieval aims at retrieving, from a collection of images, all those in which a given query object appears. It is inherently asymmetric: the query object is mostly included in the database image, while the converse is not necessarily true. However, existing approaches mostly compare the images with symmetrical measures, without considering the different roles of query and database. This paper first measure the extent of asymmetry on large-scale public datasets reflecting this task. Considering the standard bag-of-words representation, we then propose new asymmetrical dissimilarities accounting for the different inlier ratios associated with query and database images. These asymmetrical measures depend on the query, yet they are compatible with an inverted file structure, without noticeably impacting search efficiency. Our experiments show the benefit of our approach, and show that the visual object retrieval task is better treated asymmetrically, in the spirit of state-of-the-art text retrieval.

5 0.78796494 450 iccv-2013-What is the Most EfficientWay to Select Nearest Neighbor Candidates for Fast Approximate Nearest Neighbor Search?

Author: Masakazu Iwamura, Tomokazu Sato, Koichi Kise

Abstract: Approximate nearest neighbor search (ANNS) is a basic and important technique used in many tasks such as object recognition. It involves two processes: selecting nearest neighbor candidates and performing a brute-force search of these candidates. Only the former though has scope for improvement. In most existing methods, it approximates the space by quantization. It then calculates all the distances between the query and all the quantized values (e.g., clusters or bit sequences), and selects a fixed number of candidates close to the query. The performance of the method is evaluated based on accuracy as a function of the number of candidates. This evaluation seems rational but poses a serious problem; it ignores the computational cost of the process of selection. In this paper, we propose a new ANNS method that takes into account costs in the selection process. Whereas existing methods employ computationally expensive techniques such as comparative sort and heap, the proposed method does not. This realizes a significantly more efficient search. We have succeeded in reducing computation times by one-third compared with the state-of-the- art on an experiment using 100 million SIFT features.

6 0.70726871 400 iccv-2013-Stable Hyper-pooling and Query Expansion for Event Detection

7 0.68096834 221 iccv-2013-Joint Inverted Indexing

8 0.66492194 333 iccv-2013-Quantize and Conquer: A Dimensionality-Recursive Solution to Clustering, Vector Quantization, and Image Retrieval

9 0.64550817 159 iccv-2013-Fast Neighborhood Graph Search Using Cartesian Concatenation

10 0.64166266 446 iccv-2013-Visual Semantic Complex Network for Web Images

11 0.59984183 378 iccv-2013-Semantic-Aware Co-indexing for Image Retrieval

12 0.58614153 294 iccv-2013-Offline Mobile Instance Retrieval with a Small Memory Footprint

13 0.57213998 3 iccv-2013-3D Sub-query Expansion for Improving Sketch-Based Multi-view Image Retrieval

14 0.57165736 419 iccv-2013-To Aggregate or Not to aggregate: Selective Match Kernels for Image Search

15 0.51588881 235 iccv-2013-Learning Coupled Feature Spaces for Cross-Modal Matching

16 0.47467399 360 iccv-2013-Robust Subspace Clustering via Half-Quadratic Minimization

17 0.46125966 233 iccv-2013-Latent Task Adaptation with Large-Scale Hierarchies

18 0.45039836 232 iccv-2013-Latent Space Sparse Subspace Clustering

19 0.43591261 210 iccv-2013-Image Retrieval Using Textual Cues

20 0.42287099 306 iccv-2013-Paper Doll Parsing: Retrieving Similar Styles to Parse Clothing Items


similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(2, 0.038), (7, 0.01), (26, 0.028), (42, 0.706), (48, 0.01), (89, 0.105)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.96024555 162 iccv-2013-Fast Subspace Search via Grassmannian Based Hashing

Author: Xu Wang, Stefan Atev, John Wright, Gilad Lerman

Abstract: The problem of efficiently deciding which of a database of models is most similar to a given input query arises throughout modern computer vision. Motivated by applications in recognition, image retrieval and optimization, there has been significant recent interest in the variant of this problem in which the database models are linear subspaces and the input is either a point or a subspace. Current approaches to this problem have poor scaling in high dimensions, and may not guarantee sublinear query complexity. We present a new approach to approximate nearest subspace search, based on a simple, new locality sensitive hash for subspaces. Our approach allows point-tosubspace query for a database of subspaces of arbitrary dimension d, in a time that depends sublinearly on the number of subspaces in the database. The query complexity of our algorithm is linear in the ambient dimension D, allow- ing it to be directly applied to high-dimensional imagery data. Numerical experiments on model problems in image repatching and automatic face recognition confirm the advantages of our algorithm in terms of both speed and accuracy.

2 0.93601787 422 iccv-2013-Toward Guaranteed Illumination Models for Non-convex Objects

Author: Yuqian Zhang, Cun Mu, Han-Wen Kuo, John Wright

Abstract: Illumination variation remains a central challenge in object detection and recognition. Existing analyses of illumination variation typically pertain to convex, Lambertian objects, and guarantee quality of approximation in an average case sense. We show that it is possible to build models for the set of images across illumination variation with worstcase performance guarantees, for nonconvex Lambertian objects. Namely, a natural verification test based on the distance to the model guarantees to accept any image which can be sufficiently well-approximated by an image of the object under some admissible lighting condition, and guarantees to reject any image that does not have a sufficiently good approximation. These models are generated by sampling illumination directions with sufficient density, which follows from a new perturbation bound for directional illuminated images in the Lambertian model. As the number of such images required for guaranteed verification may be large, we introduce a new formulation for cone preserving dimensionality reduction, which leverages tools from sparse and low-rank decomposition to reduce the complexity, while controlling the approximation error with respect to the original model. 1

3 0.93478501 96 iccv-2013-Coupled Dictionary and Feature Space Learning with Applications to Cross-Domain Image Synthesis and Recognition

Author: De-An Huang, Yu-Chiang Frank Wang

Abstract: Cross-domain image synthesis and recognition are typically considered as two distinct tasks in the areas of computer vision and pattern recognition. Therefore, it is not clear whether approaches addressing one task can be easily generalized or extended for solving the other. In this paper, we propose a unified model for coupled dictionary and feature space learning. The proposed learning model not only observes a common feature space for associating cross-domain image data for recognition purposes, the derived feature space is able to jointly update the dictionaries in each image domain for improved representation. This is why our method can be applied to both cross-domain image synthesis and recognition problems. Experiments on a variety of synthesis and recognition tasks such as single image super-resolution, cross-view action recognition, and sketchto-photo face recognition would verify the effectiveness of our proposed learning model.

4 0.93108606 70 iccv-2013-Cascaded Shape Space Pruning for Robust Facial Landmark Detection

Author: Xiaowei Zhao, Shiguang Shan, Xiujuan Chai, Xilin Chen

Abstract: In this paper, we propose a novel cascaded face shape space pruning algorithm for robust facial landmark detection. Through progressively excluding the incorrect candidate shapes, our algorithm can accurately and efficiently achieve the globally optimal shape configuration. Specifically, individual landmark detectors are firstly applied to eliminate wrong candidates for each landmark. Then, the candidate shape space is further pruned by jointly removing incorrect shape configurations. To achieve this purpose, a discriminative structure classifier is designed to assess the candidate shape configurations. Based on the learned discriminative structure classifier, an efficient shape space pruning strategy is proposed to quickly reject most incorrect candidate shapes while preserve the true shape. The proposed algorithm is carefully evaluated on a large set of real world face images. In addition, comparison results on the publicly available BioID and LFW face databases demonstrate that our algorithm outperforms some state-of-the-art algorithms.

5 0.9130711 167 iccv-2013-Finding Causal Interactions in Video Sequences

Author: Mustafa Ayazoglu, Burak Yilmaz, Mario Sznaier, Octavia Camps

Abstract: This paper considers the problem of detecting causal interactions in video clips. Specifically, the goal is to detect whether the actions of a given target can be explained in terms of the past actions of a collection of other agents. We propose to solve this problem by recasting it into a directed graph topology identification, where each node corresponds to the observed motion of a given target, and each link indicates the presence of a causal correlation. As shown in the paper, this leads to a block-sparsification problem that can be efficiently solved using a modified Group-Lasso type approach, capable of handling missing data and outliers (due for instance to occlusion and mis-identified correspondences). Moreover, this approach also identifies time instants where the interactions between agents change, thus providing event detection capabilities. These results are illustrated with several examples involving non–trivial interactions amongst several human subjects. 1. Introduction and Motivation The problem of identifying causal interactions amongst targets in a video sequence has been the focus of considerable attention in the past few years. A large portion of the existing body of work in this field uses human annotated video to build a storyline that includes both recognizing the activities involved and the causal relationships between them (see for instance [10] and references therein). While these methods are powerful and work well when suitably annotated data is available, annotating video clips is expensive and parsing relevant actions requires domain knowledge which may not be readily available. Indeed, in many situations, unveiling potentially hidden causal relationships is a first step towards building such knowledge. In this paper we consider the problem of identifying causal interactions amongst targets, not necessarily human, ∗This work was supported by NSF grants IIS–0713003, IIS-1318145, and ECCS–0901433, AFOSR grant FA9559–12–1–0271, and the Alert DHS Center of Excellence under Award Number 2008-ST-061-ED0001 . from unannotated video sequences and without prior domain knowledge. Our approach exploits the concept of “Granger Causality” [9], that formalizes the intuitive idea that ifa time series {x(t)} is causally related to a second one {thya(tt)if}a, ttihmene knowledge }oifs tchaeu past vrealluateesd otfo a{yse}c1to should l{eya(dt t}o, a ebnett kern prediction o thf efu ptuasret vvaalulueess ooff {{yx}}tt+k. In [l1ea4d], Pora ab bheatktearr eprt.e aicl.t successfully vuasleude a frequency domain reformulation of this concept to uncover pairwise interactions in scenarios involving repeating events, such as social games. This technique was later extended in [17] to model causal correlations between human joints and applied to the problem of activity classification. However, since this approach is based upon estimating the crosscovariance density function between events, it cannot handle situations where these events are non repeating, are too rare to provide an accurate estimate, or where these estimates are biased by outliers or missing data. Further, estimating a pairwise measure of causal correlation requires a spectral factorization of the cross-covariance, followed by numerical integration and statistical thresholding, limiting the approach to moderately large problems. To circumvent these problems, in this paper we propose an alternative approach based upon recasting the problem into that of identifying the topology of a sparse (directed) graph, where each node corresponds to the time traces of relevant features of a target, and each link corresponds to a regressor. The situation is illustrated in Fig. 1 using as an example the problem of finding causal relations amongst 4 tennis players, leading to a graph with 4 nodes, and potentially 12 (directed) links. Note that in general, the problem of identifying causal relationships is ill posed (unless one wants to identify the set of all individuals that could possibly have causal connections), due to the existence of secondary interactions. To illustrate this point, consider a very simplistic scenario with three actors A, B, and C, where A copies (with some delay) the actions of B, which in turn mimics C, also with some delay. In this situation, the ac- tions of A can be explained in terms of either those of B delayed one time sample, or those of C delayed by two samples. Thus, an algorithm based upon a statistical analysis 33556758 would identify a causal connection between A and C, even though there is no direct link between them. Further, if the actions of C can be explained by some simple autoregressive model of the form: = C(t) ?aiC(t − i) then it follows that the acti?ons of A can be explained by the same model, e.g. = A(t) ?aiA(t − i) Hence, multiple graphs topologies, some of which include self-loops, can explain the same set of time-series. On the other hand, note that in this situation, the sparsest graph (in the sense of having the fewest links) is the one that correctly captures the causality relations: the most direct cause of A is B and that of B is C, with C potentially being explained by a self-loop. To capture this feature and regularize the problem, in the sequel we will seek to find the sparsest graph, in the sense of having the least number of interconnections, that explains the observed data, reflecting the fact that, when alternative models are possible, often the most parsimonious is the correct one. Our main result shows that the problem of identifying sparse graph structures from observed noisy data can be reduced to a convex optimization problem (via the use of Group Lasso type arguments) that can be efficiently solved. The advantages of the proposed methods are: • • • • Its ability to handle complex scenarios involving nonrepeating events, een cvoimropnlmeexn stcael changes, clvoillnegct nioonnsof targets that do not necessarily split into well defined groups, outliers and missing data. The ability to identify the sparsest interaction structure tThhaet explains th idee nobtifseyr tvheed s dpaartas e(stthu inst avoiding labeling as causal connections those indirect correlations mediated only by an intermediary), together with a sparse “indicator” function whose support set indicates time instants where the interactions between agents change. Since the approach is not based on semantic analysis, iSt can bt hee applied ctoh ti she n moto btiaosne dof o arbitrary targets, sniost, necessarily humans (indeed, it applies to arbitrary time series including for instance economic or genetic data). From a computational standpoint, the resulting optiFmriozmatio an c problems nhaalve s a specific fthoerm re asmuletinnagbl oep ttiobe solved by a class of iterative algorithms [5, 3], that require at each step only a combination of thresholding and least-squares approximations. These algorithms have been shown to substantially outperform conventional convex-optimization solvers both in terms of memory and computation time requirements. The remainder of the paper is organized as follows. In section 2 we provide a formal reformulation of the problem of finding causal relationships between agents as a sparse graph identification problem. In section 3, we show that this problem can be efficiently solved using a re-weighted Group Lasso approach. Moreover, as shown there, the resulting problem can be solved one node at a time using first order methods, which allows for handling situations involving a large number of agents. Finally, the effectiveness of the proposed method is illustrated in section 4 using both simple scenarios (for which ground truth is readily available) and video clips of sports, involving complex, nonrepeating interactions amongst many agents. Figure 1. Finding causal interactions as a graph identification problem. Top: sample frame from a doubles tennis sequence. Bottom: Representation of this sequence as a graph, where each node represents the time series associated with the position of each player and the links are vector regressive models. Causal interactions exist when one of the time series can be explained as a combination of past values of the others. 2. Preliminaries For ease of reference, in this section we summarize the notation used in the paper and give a formal definition of the problem under consideration. 2.1. Notation (M) ?M? ??MM??F ?M?1 ?M?o σi ∗ ◦ ith largest singular value of the matrix M. nuclear norm: ?M? ?i σ?i (M). Fnruocbleeanrio nours norm: ??M?2F? ?i,j Mi2j ?1 norm: ?M? 1 ?i,j |Mij? ?|. ?o quasi-norm: ?M?o number of non-zero ?eleme?nMts i?n M. Hadamard product of matrices: (A ◦ ∗ =.: =. =. =. B)i,j = Ai,jBi,j. 33556769 2.2. Statement of the Problem Next, we formalize the problem under consideration. Consider a scenario with P moving agents, and denote by the 3D homogenous coordinates of the pth individual at time t. Motivated by the idea of Granger Causality, we will say that the actions of this agent depend causally from those in a set Ip (which can possibly contain p itself), if can be written as: Q˜p(t) Q˜p(t) Q˜p(t) ?N = ? ?ajp(n)Q˜j(t − n) +˜ η p(t) +˜ u p(t) (1) j? ?∈Ip ?n=0 Here ajp are unknown coefficients, and ˜η p(t) and up(t) represent measurement noise and a piecewise constant signal that is intended to account for relatively rare events that cannot be explained by the (past) actions of other agents. Examples include interactions of an agent with the environment, for instance to avoid obstacles, or changes in the interactions between agents. Since these events are infrequent, we will model as a signal that has (component-wise) a sparse derivative. Note in passing that since (1) involves homogeneous coordinates, the coefficients aj,p(.) satisfy the following constraint1 u ?N ? ?ajp(n) j? ?∈Ip ?n=0 =1 (2) Our goal is to identify causal relationships using as data 2D measurements qp(t) in F frames of the affine projections of the 3D coordinates Q˜p(t) of the targets. Note that, under the affine camera assumption, the 2D coordinates are related exactly by the same regressor parameters [2]. Thus, (1) holds if and only if: ?N qp(t) = ? ?ajp(n)qj(t − n) + u˜ p(t) + ηp(t) (3) j?∈Ip ?n=0 In this context, the problem can be precisely stated as: Given qp(t) (in F number of frames) and some a-priori bound N on the order of the regressors (that is the “memory” of the interactions), find the sparsest set of equations of the form (3) that explains the data, that is: aj,pm,ηinp,up?nIp (4) subject to? ?(2) and: = ? ?ajp(n)qj(t − n) + ?N qp(t) j? ?∈Ip ?n=0 up(t) + ηp(t) , p = 1 . . . , P and t = 1, ..F 1This follows by considering the third coordinate in (1) (5) where nIp denotes the cardinality of the set Ip. Rewriting (5) in matrixd efnoormtes yields: [xp; yp] = [Bp, I][apTuxTpuyTp]T + ηp (6) where qp(t) up(t) ηp(t) xp yp ap aip uxp uyp Bp Xp = [xp(t)Typ(t)T]T = [uTxp(t)uyTp(t)]T = [ηxp(t)Tηyp(t)T]T = = [xp(F)xp(F − 1)...xp(1)]T = [yp(F)yp(F − 1)...yp(1)]T [aT1p, a2Tp, ..., aTPp]T = [aip(0), aip(1), ..., aip(N)]T = [uxp(F)uxp(F−1)...uxp(1)]T = [uyp(F)uyp(F−1)...uyp(1)]T = = [Xp; Yp] [hankel(x1 , N) , ..., hankel(xP, N)] Yp = [hankel(y1, N), ..., hankel(yP, N)] and where, for a sequence z(t), hankel(z, N) denotes its associated Hankel matrix: hankel(z, N) = Itfolw⎛⎜⎝ sz t(hNzFa(t. +−a)d1 2e)scrzip(tF io(N. n− )o231f)al· t h· einzt(Frac−zti(.o1N.n)s−a)m12o)⎟ ⎞⎠ ngst uηaq= ? ηuqa1 T ,ηqau2 T ,ηaqu3 T ,· ·, ηauqP T ? T (8) Thus,inthBisc=on⎢⎣⎡teBx0t.1,theB0p.r2ob·le.·m·ofB0 i.nPte⎦⎥r ⎤estcanbeforagents (that is the complete graph structure) is captured by a matrix equation of the form: q = [B, I][aTuT]T + η (7) where and malized as finding the block–sparsest solution to the set of linear equations (2) and (7). 33557770 The problem of identifying a graph structure subject to sparsity constraints, has been the subject of intense research in the past few years. For instance, [1] proposed a Lasso type algorithm to identify a sparse network where each link corresponds to a VAR process. The main idea underlying this method is to exploit the fact that penalizing the ?1 norm of the vector of regression coefficients tends to produce sparse solutions. However, enforcing sparsity of the entire vector of regressor coefficients does not necessarily result in a sparse graph structure, since the resulting solution can consist of many links, each with a few coefficients. This difficulty can be circumvented by resorting to group Lasso type approaches [18], which seek to enforce block sparsity by using a combination of ?1 and ?2 norm constraints on the coefficients of the regressor. While this approach was shown to work well with artificial data in [11], exact recovery of the underlying network can be only guaranteed when the data satisfies suitable “incoherence” type conditions [4]. Finally, a different approach was pursued in [13], based on the use of a modified Orthogonal Least Squares algorithm, Cyclic Orthogonal Least Squares. However, this approach requires enforcing an a-priori limit on the number of links allowed to point to a single node, and such information may not be readily available, specially in cases where this number has high variability amongst nodes. To address these difficulties, in the next section we develop a convex optimization based approach to the problem of identifying sparse graph structures from observed noisy data. This method is closest in spirit to that in [11], in the sense that it is also based on a group Lasso type argument. The main differences consist in the ability to handle the unknown inputs up(t), needed to model exogenous disturbances affecting the agents, and in a reformulation of the problem, that allows for using a re-weighted iterative type algorithm, leading to substantially sparser solutions, even when the conditions in [4] fail. 3. Causality Identification Algorithm In this section we present the main result of this paper, an algorithm to search for block-sparse solutions to (7). For each fixed p, the algorithm searches for sparse solutions to (6) by solving (iteratively) the following problem (suggested by the re-weighted heuristic proposed in [7]) ?P ap,muxipn,uypi?=1wja(?aip?2) + λ??diag(wu)[Δuxp;Δuyp]??1 subject to: ?ηp ? ≤ p = 1, . . , P. ∞ ?P ?, ?N ??aip(n) i?= ?1 ?n=0 ?. = 1, p = 1,...,P. (9) where [Δuxp ; Δuyp] represents the first order differences of the exogenous input vector [uxp ; uyp], Wa and Wu are weighting matrices, and λ is a Lagrange multiplier that plays the role of a tuning parameter between graph sparsity and event sensitivity. Intuitively, for a fixed set of weights w, the algorithm attempts to find a block sparse solution to (6) and a set of sparse inp?uts Δuxp ; Δuyp , by exploiting the facts that minimizing ?i ?aip ?2 (the ?2,1 norm of the vector sequence {aip}) te?nds? tao m?aximize block-sparsity [18], while minimizing et?nhed s? 1t norm mmaizxeim blizoceks sparsity [ [1168]]. wOhniclee t mheisnesolutions are found, the weights w are adjusted to penalize those elements of the sequences with small values, so that in the next iteration solutions that set these elements to zero (hence further increasing sparsity) are favored. Note however, that proceeding in this way, requires solving at each iteration a problem with n = P(Pnr + F) variables, where P and F denote the number of agents and frames, respectively, and where nr is a bound on the regressor order. On the other hand, it is easily seen that both the objective function and the constraints in (9) can be partitioned into P groups, with the pth group involving only the variables related to the pth node. It follows then that problem (9) can be solved by solving P smaller problems of the form: ?P ap,muxipn,uypi?=1wja(?aip?2) + λ??diag(wu)[Δuxp;Δuyp]??1 ?P subject to: ?ηp?∞ ?N ≤ ? and ??aip(n) i?= ?1 ?n=0 leading to the algorithm given below: =1 (10) Algorithm 1: REWEIGHTEDCAUSALITYALGORITHM for each p wa = [1, 1, ..., 1] = [1, 1, ..., 1] S > 1(self loop weight) s = [1, 1, ..., S, ..., 1] (p’th element is S) while not converged do 1. solve (9) 2. wja = 1/( ?aip ?2 + δ) 3. wja = wja ◦ s (Penalization self loops) 4. = 1./(abs([Δuxp ; Δuyp]) + δ) end while 5. At this point ajp(.) , Ip and up(t) have been identified end for wu wu It is worth emphasizing that, since the computational complexity of standard interior point methods grows as n3, solving these smaller P problems leads to roughly a O(P2) 33557781 reduction in computational time over solving a single, larger optimization. Thus, this approach can handle moderately large problems using standard, interior-point based, semidefinite optimization solvers. Larger problems can be accommodated by noting that the special form of the objective and constraints allow for using iterative Augmented La- grangian Type Methods (ALM), based upon computing, at each step, the closed form solution to suitable intermediate optimization problems. While a complete derivation of such an algorithm is beyond the scope of this paper, using results from [12] it can be shown that each step requires only a combination of thresholding and least-squares approximations. Moreover, it can be shown that such an algorithm converges Q-superlinearly. 4. Handling Outliers and Missing Data The algorithm outlined above assumes an ideal situation where the data matrix B is perfectly known. However, in practice many of its elements may be outliers (due to misidentified correspondences) or missing (due to occlusion). As we briefly show next, these situations can be efficiently handled by performing a structured robust PCA step [3] to obtain a “clean” data matrix, prior to applying Algorithm 1. From equation (6) it follows that, in the absence of exogenous inputs and noise: ?xy11.. . .yxPP? = ?XY11.. . .YXPP? ?a1...aP? (11) Since xi ∈ {col(Xj)} and yi ∈ {col(Yj }), it follows that the sets {∈co {l(cXoli(X)} a)n}d a n{dco yl(Y∈i) { }c? are self-ex?pressive, or, ?equivalently?, Xthe }ma atnridce {sc oXl( =.) }? aXre1 . . . fX-eNxp? eanssdiv eY, ?Y1 ...YN? are mraantkri cdeesfic Xient. ?Consider no?w the case =.r, w?here some ?elements xi, yi of X and Y are missing. From ?the self-expressive property ooff {Xco aln(Xd Yi)} a raen dm i{scsoinlg(Y. Fi)ro} mit tfhoello swelsf tehxaptr ethsessieve missing eyle omf {encotsl are given by: xi = argmin rank(X) , yi x = argmin rank(Y) (12) y Similarly, in the presence of outliers, X, Y can be decomposed irnlyto, itnhe t sum oesfe a lcoew o fra onkut mlieartsr,ix X (,thYe ccalenan b eda dtae)c oamnda sparse one (the outliers) by solving a problem of the form minrank?YXoo?+ λ????EEYX????os. t.: ?XYoo?+?EEYX?=?YX? From the reasoning? abov?e it follows that in the presence of noise and exogenous outputs, the clean data record can be recovered from the corrupted, partial measurements by solving the following optimization problem: s+muλibn3je? ? ? cYXtM ot ? Y?X:∗◦ +Ξ λYX1? ? ? FM XY◦ E XY? ?1+λ2? ?M YX◦ Δ U YX? ?1 ?YX?=?XYoo?+?EEXY?+?UUYX?+?ΞΞYX? (13) where we have used the standard convex relaxations of rank and cardinality2. Here Ξ and U denote noise and piecewise constant exogenous matrices, ΔU denotes the matrix obtained by taking the difference between consecutive elements in U, and MX (MY) is a “mask” matrix, with mi,j = 0 if the element (i, j) in X ( Y) is missing, mi,j = 1 otherw=i0s e, i tuhseed e etom aenvtoi (di, penalizing )e lisem miesnstisn gin, mE, Ξ, U corresponding to missing data. Problem (13) is a structured robust PCA problem (due to the Hankel structure of X, Y) trhobatu can C bAe efficiently suoelv teod t using tkheel fsitrrsut oturrdeer o mf Xeth,oYd) proposed in [3], slightly modified to handle the terms containing ΔU. 5. Experimental Results In this section we illustrate the effectiveness of the proposed approach using several video clips (provided as supplemental material). The results of the experiments are displayed using graphs embedded on the video frames: An arrow indicates causal correlation between agents, with the point of the arrow indicating the agent whose actions are affected by the agent at its tail. The internal parameters of the algorithm were experimentally tuned, leading to the values ? = 0.1, = 0.05, self loop weights S = 10. The algorithm is fairly insensitive to the value of the regularization parameters and S, which could be adjusted up or down by an order of magnitude without affecting the structure of the resulting graph. Finally, we used regressor order N=2 for the first three examples and N=4 for the last one, a choice that is consistent with the frame rate and the complexity of λ λ the actions taking place in each clip. 5.1. Clips from the UT-Interaction Data Set We considered two video clips from the UT Human Interaction Data Set [15] (sequences 6 and 16). Figures 2 and 5 compare the results obtained applying the proposed algorithm versus Group Lasso (GL) [11] and Group Lasso combined with the reweighted heuristic described in (9) (GLRW). In all cases, the inputs to the algorithm were the (approximate) coordinates of the heads of each of the agents, normalized to the interval [−1, 1], artificially corrupted ,w niothrm m10al%iz eodut tloie trhs.e Notably, [t−he1 proposed algorithm 2As shown in [6, 8] under suitable conditions these relaxations the exact minimum rank solution. 33557792 recover Figure 2. Sample frames from the UT sequence 6 with the identified causal connections superimposed. Top: Proposed Method. Center: Reweighted Group Lasso. Bottom: Group Lasso. Only the proposed method identifies the correct connections. was able to correctly identify the correlations between the agents from this very limited amount of information, while the others failed to do so. Note in passing that in both cases none of the algorithms were directly applicable, due to some of the individuals leaving the field of view or being occluded. As illustrated in Fig. 3, the missing data was recovered by solving an RPCA problem prior to applying Algorithm 1. Finally, Fig. 4 sheds more insight on the key role played by the sparse signal u. As shown there, changes in u correspond exactly to time instants when the behavior of the corresponding agent deviates from the general pattern followed during most of the clip. Figure 3. Time traces of the individual heads in the UT sequence 6, artificially corrupted with 10 % outliers. The outliers were removed and the missing data due to targets leaving the field of view was estimated solving a modified RPCA problem. Frame number Figure 4. Sample (derivative sparse) exogenous signals in the UT sequence 6. The changes correspond to the instants when the second person starts moving towards the first, who remains stationary, and when the two persons merge in an embrace. Figure 5. Sample frames from the UT sequence 16. Top: Correct correlations identified by the Proposed Method. Center and Bottom: Reweighted Group Lasso and Group Lasso (circles indicate self-loops). 5.2. Doubles Tennis Experiment This experiment considers a non-staged real-life scenario. The data consists of 230 frames of a video clip from the Australian Open Tennis Doubles Final games. The goal here is to identify causal relationships between the different players using time traces of the respective centroid positions. Note that in this case the ground truth is not available. Nevertheless, since players from the same team usually look at their opponents and react to their motions, we expect a strong causality connection between members of 33557803 opposite teams. This intuition is matched by the correlations unveiled by the algorithm, shown in Fig. 6. The identified sparse input corresponding to the vertical direction is shown in Fig. 7 (similar results for the horizontal component are omitted due to space reasons.) Figure 6. Sample frames from the tennis sequence. Top: The proposed method correctly identifies interactions between opposite team members. Center: Reweighted Group Lasso misses the interaction between the two rear-most individuals of opposite teams, generating self loops instead (denoted by the disks). Bottom: Group Lasso yields an almost complete graph. Figure 7. Exogenous signal corresponding to the vertical axis for the tennis sequence. The change in one component corresponds to the instant when the leftmost player in the bottom team moves from the line towards the net, remaining closer to it from then on. 5.3. Basketball Game Experiment This experiment considers the interactions amongst players in a basketball game. As in the case ofthe tennis players, since the data comes from a real life scenario, the ground truth is not available. However, contrary to the tennis game, this scenario involves complex interactions amongst many players, and causality is hard to discern by inspection. Nevertheless, the results shown in Fig. 8, obtained using the position of the centroids as inputs to our algorithm, match our intuition. Firstly, one would expect a strong cause/effect connection between the actions of the player with the ball and the two defending opponents facing him. These connections (denoted by the yellow arrows) were indeed successfully identified by the algorithm. The next set of causal correlations is represented by the (blue, light green) and (black, white) arrow pairs showing the defending and the opponent players on the far side of the field and under the hoop. An important, counterintuitive, connection identified by the algorithm is represented by the magenta arrows be- tween the right winger of the white team with two of his teammates: the one holding the ball and the one running behind all players. While at first sight this connection is not as obvious as the others, it becomes apparent towards the end of the sequence, when the right winger player is signaling with a raised arm. Notably, our algorithm was able to unveil this signaling without the need to perform a semantic analysis (a very difficult task here, since this signaling is apparent only in the last few frames). Rather, it used the fact that the causal correlation was encapsulated in the dynamics of the relative motions of these players. 6. Conclusions In this paper we propose a new method for detecting causal interactions between agents using video data. The main idea is to recast this problem into a blind directed graph topology identification, where each node corresponds to the observed motion of a given target, each link indicates the presence of a causal correlation and the unknown inputs account for changes in the interaction patterns. In turn, this problem can be reduced to that of finding block-sparse solutions to a set of linear equations, which can be efficiently accomplished using an iterative re-weighted Group-Lasso approach. The ability of the algorithm to correctly identify causal correlations, even in cases where portions of the data record are missing or corrupted by outliers, and the key role played by the unknown exogenous input were illustrated with several examples involving non–trivial inter- actions amongst several human subjects. Remarkably, the proposed algorithm was able to identify both the correct interactions and the time instants when interactions amongst agents changed, based on minimal motion information: in all cases we used just a single time trace per person. This success indicates that in many scenarios, the dynamic information contained in the motion pattern of a single feature associated with a target is rich enough to enable identifying complex interaction patterns, without the need to track multiple features, perform a semantic analysis or use additional domain knowledge. 33557814 Figure 8. Sample frames from a Basketball game. Top: proposed method. Center: Reweighted Group the signaling player and his teammates. Bottom: Group Lasso yields an almost complete graph. Lasso misses the interaction between References [1] A. Arnold, Y. Liu, and N. Abe. Estimating brain functional connectivity with sparse multivariate autoregression. In Proc. of the 13th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pages 66–75, 2007. 4 [2] M. Ayazoglu, B. Li, C. Dicle, M. Sznaier, and O. Camps. Dynamic subspace-based coordinated multicamera tracking. In 2011 IEEE ICCV, pages 2462–2469, 2011. 3 [3] M. Ayazoglu, M. Sznaier, and O. Camps. Fast algorithms for structured robust principal component analysis. In 2012 IEEE CVPR, pages 1704–171 1, June 2012. 2, 5 [4] A. Bolstad, B. Van Veen, and R. Nowak. Causal network inference via group sparse regularization. IEEE Transactions on Signal Processing, 59(6):2628–2641, 2011. 4 [5] S. Boyd, N. Parikh, E. Chu, B. Peleato, and J. Eckstein. Dis- [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] tributed optimization and statistical learning via the alternating direction method of multipliers. Found. Trends Mach. Learn., 3(1):1–122, Jan. 2011. 2 E. Candes, X. Li, Y. Ma, and J.Wright. Robust principal component analysis? J. ACM, (3), 2011. 5 E. J. Candes, M. Wakin, and S. Boyd. Enhancing sparsity by reweighted l1minimization. Journal of Fourier Analysis and Applications, 14(5):877–905, December 2008. 4 V. Chandrasekaran, S. Sanghavi, P. Parrilo, and A. Willsky. Rank-sparsity incoherence for matrix decomposition. Siam J. Optim., (2):572–596, 2011. 5 C. W. J. Granger. Investigating causal relations by econometric models and cross-spectral methods. Econometrica, pages 424–438l, 1969. 1 A. Gupta, P. Srinivasan, J. Shi, and L. Davis. Understanding videos, constructing plots: Learning a visually grounded storyline model from annotated videos. In 2009 IEEE CVPR, pages 2012–2019, 2009. 1 S. Haufe, G. Nolte, K. R. Muller, and N. Kramer. Sparse causal discovery in multivariate time series. In Neural Information Processing Systems, 2009. 4, 5 G. Liu, Z. Lin, and Y. Yu. Robust subspace segmentation by low-rank representation. In ICML, pages 1663–670, 2010. 5 D. Materassi, G. Innocenti, and L. Giarre. Reduced complexity models in identification of dynamical networks: Links with sparsification problems. In 48th IEEE Conference on Decision and Control, pages 4796–4801, 2009. 4 K. Prabhakar, S. Oh, P. Wang, G. Abowd, and J. Rehg. Temporal causality for the analysis ofvisual events. In IEEE Conf Comp. Vision and Pattern Recog. (CVPR)., pages 1967– 1974, 2010. 1 M. S. Ryoo and J. K. Aggarwal. UT Interaction Dataset, ICPR contest on Semantic Description of Human Activities. http://cvrc.ece.utexas.edu/SDHA2010/Human Interaction.html, 2010. 5 [16] J. Tropp. Just relax: convex programming methods for identifying sparse signals in noise. IEEE Transactions on Information Theory, 52(3): 1030–1051, 2006. 4 [17] S. Yi and V. Pavlovic. Sparse granger causality graphs for human action classification. In 2012 ICPR, pages 3374–3377. 1 [18] M. Yuan and Y. Lin. Model selection and estimation in regression with grouped variables. Journal of the Royal Statistical Society Series B, 68(1):49–67, 2006. 4 33557825

6 0.89648521 213 iccv-2013-Implied Feedback: Learning Nuances of User Behavior in Image Search

7 0.86953467 46 iccv-2013-Allocentric Pose Estimation

8 0.85057169 184 iccv-2013-Global Fusion of Relative Motions for Robust, Accurate and Scalable Structure from Motion

9 0.82018214 231 iccv-2013-Latent Multitask Learning for View-Invariant Action Recognition

10 0.79026783 93 iccv-2013-Correlation Adaptive Subspace Segmentation by Trace Lasso

11 0.78360403 14 iccv-2013-A Generalized Iterated Shrinkage Algorithm for Non-convex Sparse Coding

12 0.77126497 54 iccv-2013-Attribute Pivots for Guiding Relevance Feedback in Image Search

13 0.74473506 161 iccv-2013-Fast Sparsity-Based Orthogonal Dictionary Learning for Image Restoration

14 0.73181558 398 iccv-2013-Sparse Variation Dictionary Learning for Face Recognition with a Single Training Sample per Person

15 0.73104161 259 iccv-2013-Manifold Based Face Synthesis from Sparse Samples

16 0.7253868 114 iccv-2013-Dictionary Learning and Sparse Coding on Grassmann Manifolds: An Extrinsic Solution

17 0.72246641 154 iccv-2013-Face Recognition via Archetype Hull Ranking

18 0.7195853 52 iccv-2013-Attribute Adaptation for Personalized Image Search

19 0.71524191 335 iccv-2013-Random Faces Guided Sparse Many-to-One Encoder for Pose-Invariant Face Recognition

20 0.71439642 106 iccv-2013-Deep Learning Identity-Preserving Face Space