Results 1 - 10
of
60
Person reidentification by salience matching
- In ICCV
, 2013
"... Human salience is distinctive and reliable information in matching pedestrians across disjoint camera views. In this paper, we exploit the pairwise salience distribution re-lationship between pedestrian images, and solve the person re-identification problem by proposing a salience matching strategy. ..."
Abstract
-
Cited by 25 (3 self)
- Add to MetaCart
(Show Context)
Human salience is distinctive and reliable information in matching pedestrians across disjoint camera views. In this paper, we exploit the pairwise salience distribution re-lationship between pedestrian images, and solve the person re-identification problem by proposing a salience matching strategy. To handle the misalignment problem in pedes-trian images, patch matching is adopted and patch salience is estimated. Matching patches with inconsistent salience brings penalty. Images of the same person are recognized by minimizing the salience matching cost. Furthermore, our salience matching is tightly integrated with patch match-ing in a unified structural RankSVM learning framework. The effectiveness of our approach is validated on the VIPeR dataset and the CUHK Campus dataset. It outperforms the state-of-the-art methods on both datasets. 1.
Deepreid: Deep filter pairing neural network for person re-identification
- In CVPR
, 2014
"... Person re-identification is to match pedestrian images from disjoint camera views detected by pedestrian detec-tors. Challenges are presented in the form of complex varia-tions of lightings, poses, viewpoints, blurring effects, image resolutions, camera settings, occlusions and background clutter ac ..."
Abstract
-
Cited by 17 (3 self)
- Add to MetaCart
(Show Context)
Person re-identification is to match pedestrian images from disjoint camera views detected by pedestrian detec-tors. Challenges are presented in the form of complex varia-tions of lightings, poses, viewpoints, blurring effects, image resolutions, camera settings, occlusions and background clutter across camera views. In addition, misalignment in-troduced by the pedestrian detector will affect most existing person re-identification methods that use manually cropped pedestrian images and assume perfect detection. In this paper, we propose a novel filter pairing neural network (FPNN) to jointly handle misalignment, photomet-ric and geometric transforms, occlusions and background clutter. All the key components are jointly optimized to maximize the strength of each component when cooperat-ing with others. In contrast to existing works that use hand-crafted features, our method automatically learns features optimal for the re-identification task from data. The learned filter pairs encode photometric transforms. Its deep archi-tecture makes it possible to model a mixture of complex photometric and geometric transforms. We build the largest benchmark re-id dataset with 13,164 images of 1,360 pedes-trians. Unlike existing datasets, which only provide manu-ally cropped pedestrian images, our dataset provides au-tomatically detected bounding boxes for evaluation close to practical applications. Our neural network significantly outperforms state-of-the-art methods on this dataset. 1.
LEARNING SIMILARITY FROM COLLABORATIVE FILTERS
"... Collaborative filtering methods (CF) exploit the wisdom of crowds to capture deeply structured similarities in musical objects, such as songs, artists or albums. When CF is available, it frequently outperforms content-based methods in recommendation tasks. However, songs in the so-called “long tail ..."
Abstract
-
Cited by 16 (1 self)
- Add to MetaCart
(Show Context)
Collaborative filtering methods (CF) exploit the wisdom of crowds to capture deeply structured similarities in musical objects, such as songs, artists or albums. When CF is available, it frequently outperforms content-based methods in recommendation tasks. However, songs in the so-called “long tail ” cannot reap the benefits of collaborative filtering, and practitioners must rely on content-based methods. We propose a method for improving contentbased recommendation in the long tail by learning an optimized similarity function from a sample of collaborative filtering data. Our experimental results demonstrate substantial improvements in accuracy by learning optimal similarity functions. 1.
Set based discriminative ranking for recognition
- in ECCV, 2012
"... Abstract. Recently both face recognition and body-based person re-identification have been extended from single-image based scenarios to video-based or even more generally image-set based problems. Set-based recognition brings new research and application opportunities while at the same time raises ..."
Abstract
-
Cited by 15 (5 self)
- Add to MetaCart
(Show Context)
Abstract. Recently both face recognition and body-based person re-identification have been extended from single-image based scenarios to video-based or even more generally image-set based problems. Set-based recognition brings new research and application opportunities while at the same time raises great modeling and optimization challenges. How to make the best use of the available multiple samples for each individual while at the same time not be disturbed by the great within-set varia-tions is considered by us to be the major issue. Due to the difficulty of designing a global optimal learning model, most existing solutions are still based on unsupervised matching, which can be further categorized into three groups: a) set-based signature generation, b) direct set-to-set matching, and c) between-set distance finding. The first two count on good feature representation while the third explores data set structure and set-based distance measurement. The main shortage of them is the lack of learning-based discrimination ability. In this paper, we propose a set-based discriminative ranking model (SBDR), which iterates between set-to-set distance finding and discriminative feature space projection to achieve simultaneous optimization of these two. Extensive experiments on widely-used face recognition and person re-identification datasets not only demonstrate the superiority of our approach, but also shed some light on its properties and application domain. 1
Playlist prediction via metric embedding
- In Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining
, 2012
"... Digital storage of personal music collections and cloud-based music services (e.g. Pandora, Spotify) have fundamentally changed how music is consumed. In particular, automati-cally generated playlists have become an important mode of accessing large music collections. The key goal of automated playl ..."
Abstract
-
Cited by 14 (4 self)
- Add to MetaCart
(Show Context)
Digital storage of personal music collections and cloud-based music services (e.g. Pandora, Spotify) have fundamentally changed how music is consumed. In particular, automati-cally generated playlists have become an important mode of accessing large music collections. The key goal of automated playlist generation is to provide the user with a coherent lis-tening experience. In this paper, we present Latent Markov Embedding (LME), a machine learning algorithm for gen-erating such playlists. In analogy to matrix factorization methods for collaborative filtering, the algorithm does not require songs to be described by features a priori, but it learns a representation from example playlists. We formu-late this problem as a regularized maximum-likelihood em-bedding of Markov chains in Euclidian space, and show how the resulting optimization problem can be solved efficiently. An empirical evaluation shows that the LME is substantially more accurate than adaptations of smoothed n-gram models commonly used in natural language processing.
Learning Hierarchical Similarity Metrics
"... Categories in multi-class data are often part of an underlying semantic taxonomy. Recent work in object classification has found interesting ways to use this taxonomy structure to develop better recognition algorithms. Here we propose a novel framework to learn similarity metrics using the class tax ..."
Abstract
-
Cited by 14 (0 self)
- Add to MetaCart
(Show Context)
Categories in multi-class data are often part of an underlying semantic taxonomy. Recent work in object classification has found interesting ways to use this taxonomy structure to develop better recognition algorithms. Here we propose a novel framework to learn similarity metrics using the class taxonomy. We show that a nearest neighbor classifier using the learned metrics gets improved performance over the best discriminative methods. Moreover, by incorporating the taxonomy, our learned metrics can also help in some taxonomy specific applications. We show that the metrics can help determine the correct placement of a new category that was not part of the original taxonomy, and can provide effective classification amongst categories local to specific subtrees of the taxonomy. 1.
Metric Learning with Multiple Kernels
"... Metric learning has become a very active research field. The most popular representative–Mahalanobis metric learning–can be seen as learning a linear transformation and then computing the Euclidean metric in the transformed space. Since a linear transformation might not always be appropriate for a g ..."
Abstract
-
Cited by 13 (4 self)
- Add to MetaCart
(Show Context)
Metric learning has become a very active research field. The most popular representative–Mahalanobis metric learning–can be seen as learning a linear transformation and then computing the Euclidean metric in the transformed space. Since a linear transformation might not always be appropriate for a given learning problem, kernelized versions of various metric learning algorithms exist. However, the problem then becomes finding the appropriate kernel function. Multiple kernel learning addresses this limitation by learning a linear combination of a number of predefined kernels; this approach can be also readily used in the context of multiple-source learning to fuse different data sources. Surprisingly, and despite the extensive work on multiple kernel learning for SVMs, there has been no work in the area of metric learning with multiple kernel learning. In this paper we fill this gap and present a general approach for metric learning with multiple kernel learning. Our approach can be instantiated with different metric learning algorithms provided that they satisfy some constraints. Experimental evidence suggests that our approach outperforms metric learning with an unweighted kernel combination and metric learning with cross-validation based kernel selection. 1
Robust Structural Metric Learning
"... Metric learning algorithms produce a linear transformation of data which is optimized for a prediction task, such as nearest-neighbor classification or ranking. However, when the input data contains a large portion of noninformative features, existing methods fail to identify the relevant features, ..."
Abstract
-
Cited by 12 (1 self)
- Add to MetaCart
(Show Context)
Metric learning algorithms produce a linear transformation of data which is optimized for a prediction task, such as nearest-neighbor classification or ranking. However, when the input data contains a large portion of noninformative features, existing methods fail to identify the relevant features, and performance degrades accordingly. In this paper, we present an efficient and robust structural metric learning algorithm which enforces group sparsity on the learned transformation, while optimizing for structured ranking output prediction. Experiments on synthetic and real datasets demonstrate that the proposed method outperforms previous methods in both high- and low-noise settings. 1.
Optimizing mean reciprocal rank for person re-identification
- In AVSS
, 2011
"... Person re-identification is one of the most challenging is-sues in network-based surveillance. The difficulties mainly come from the great appearance variations induced by il-lumination, camera view and body pose changes. Maybe influenced by the research on face recognition and gen-eral object recog ..."
Abstract
-
Cited by 11 (6 self)
- Add to MetaCart
(Show Context)
Person re-identification is one of the most challenging is-sues in network-based surveillance. The difficulties mainly come from the great appearance variations induced by il-lumination, camera view and body pose changes. Maybe influenced by the research on face recognition and gen-eral object recognition, this problem is habitually treated as a verification or classification problem, and much effort has been put on optimizing standard recognition criteria. However, we found that in practical applications the users usually have different expectations. For example, in a real surveillance system, we may expect that a visual user inter-face can show us the relevant images in the first few (e.g. 20) candidates, but not necessarily before all the irrelevant ones. In other words, there is no problem to leave the fi-nal judgement to the users. Based on such an observation, this paper treats the re-identification problem as a ranking problem and directly optimizes a listwise ranking function named Mean Reciprocal Rank (MRR), which is considered by us to be able to generate results closest to human expec-tations. Using a maximum-margin based structured learn-ing model, we are able to show improved re-identification results on widely-used benchmark datasets. 1.
