Normalized Cuts and Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Machine Learning
Viewpoint Invariant Pedestrian Recognition with an Ensemble of Localized Features
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I
Learning Discriminative Appearance-Based Models Using Partial Least Squares
SIBGRAPI '09 Proceedings of the 2009 XXII Brazilian Symposium on Computer Graphics and Image Processing
Time-Delayed Correlation Analysis for Multi-Camera Activity Understanding
International Journal of Computer Vision
Multiple-shot person re-identification by chromatic and epitomic analyses
Pattern Recognition Letters
Foundations and Trends® in Computer Graphics and Vision
Person re-identification by probabilistic relative distance comparison
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
PCCA: A new approach for distance learning from sparse pairwise constraints
CVPR '12 Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Fast person re-identification based on dissimilarity representations
Pattern Recognition Letters
People reidentification in surveillance and forensics: A survey
ACM Computing Surveys (CSUR)
On-the-fly feature importance mining for person re-identification
Pattern Recognition
Editor's Choice Article: A survey of approaches and trends in person re-identification
Image and Vision Computing
Hi-index | 0.00 |
State-of-the-art person re-identification methods seek robust person matching through combining various feature types. Often, these features are implicitly assigned with a single vector of global weights, which are assumed to be universally good for all individuals, independent to their different appearances. In this study, we show that certain features play more important role than others under different circumstances. Consequently, we propose a novel unsupervised approach for learning a bottom-up feature importance, so features extracted from different individuals are weighted adaptively driven by their unique and inherent appearance attributes. Extensive experiments on two public datasets demonstrate that attribute-sensitive feature importance facilitates more accurate person matching when it is fused together with global weights obtained using existing methods.