Efficient PSD Constrained Asymmetric Metric Learning for Person Re-Identification

By Shengcai Liao, S. Li. IEEE International Conference on Computer Vision, 2015

404 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/6fd3fd7840f9995366a687569b2ece97c66cf336

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Liang Zheng
2. Yi Yang
3. Qi Tian
4. Shaogang Gong
5. Tao A. Xiang
6. Shiliang Zhang
7. Wanli Ouyang
8. Mang Ye

In Computer Vision and Pattern Recognition

1. Yi Yang
2. Liang Zheng
3. Xiaogang Wang
4. Luc Van Gool
5. Dacheng Tao
6. Qi Tian
7. Ming-Hsuan Yang
8. Tao A. Xiang

Citations in the corpus

• Deeply-Learned Part-Aligned Representations for Person Re-identification (2017, methodology): “A lot of similarity/metric learning techniques [57, 58, 33, 27, 19] have been applied or designed to learn metrics, robust to light/view/pose changes, for person matching.”
• High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identification (2020, methodology): “Existing methods can be grouped into hand-crafted descriptors [25, 39, 18], metric learning methods [44, 16, 19] and deep learning algorithms [43, 11, 35, 36, 37, 22].”
• Learning Deep Context-Aware Features over Body and Latent Parts for Person Re-identification (2017, methodology): “CUHK03: For the CUHK03 dataset, we compare our method with many existing approaches, including Filter Pair Neural Networks (FPNN) [20], Improved Deep Learning Architecture (IDLA) [1], Cross-view Quadratic Discriminant Analysis (XQDA) [22], PSD constrained asymmetric metric learning (denoted as MLAPG”
• Enhancing Person Re-identification in a Self-Trained Subspace (2017, methodology): “[16] proposed a logistic metric learning approach with PSD constraints and asymmetric sample weight strategy.”
• Bi-Directional Center-Constrained Top-Ranking for Visible Thermal Person Re-Identification (2020, background): “In addition, some matching model learning methods (XQDA [23], MLAPG [59], GSM [58] SCDL [60] and rCDL [61]) are also included for comparison.”
• Learning to Reduce Dual-Level Discrepancy for Infrared-Visible Person Re-Identification (2019, methodology): “In addition, two matching model learning methods, MLAPG [9] and GSM [10], are also included for comparisons.”
• Gated Siamese Convolutional Neural Network Architecture for Human Re-identification (2016, methodology): “Several Metric Learning algorithms such as Locally adaptive Decision Functions (LADF) [23], Cross-view Quadratic Discriminant Analysis (XQDA) [24], Metric Learning with Accelerated Proximal Gradient (MLAPG) [25], Local Fisher Discriminant Analysis (LFDA) [32] and its kernel variant (k-LFDA) [49] wer”
• A Siamese Long Short-Term Memory Architecture for Human Re-identification (2016, neutral): “MLAPG [32] 51.”
• Beyond Triplet Loss: A Deep Quadruplet Network for Person Re-identification (2017, methodology): “Most of existing methods in person ReID focus on either feature extraction [44, 28, 23, 8, 40], or similarity measurement [17, 27, 22, 39].”
• GLAD: Global-Local-Alignment Descriptor for Pedestrian Retrieval (2017, methodology): “On CUHK03, we compare GLAD with recent distance metric learning based methods, including WARCA [15], LOMO + XQDA [18], Null Space [41] and MLAPG [19].”
• Hierarchical Gaussian Descriptor for Person Re-identification (2016, methodology): “5% rank-1 identification rates with the labeled and the automatically detected bounding boxes, respectively, which clearly outperforms the state-of-the-art LOMO features [23, 24] and deep learning methods [1, 21] by a large margin.”
• Deep Transfer Learning for Person Re-Identification (2016, background): “Earlier works focus on either designing viewinsensitive feature representations [2], [3], or learning an effective distance metric [4], [5], or both [6].”
• End-to-End Comparative Attention Networks for Person Re-Identification (2016, methodology): “Theoretic Metric Learning (ITML) [1], Kernelized Relaxed Margin Components Analysis (KRMCA) [45], Robust Canonical Correlation Analysis (ROCCA) [21], Metric Learning with Accelerated Proximal Gradient (MLAPG) [31] and transfer Local Relative Distance Comparison (t-LRDC) are represen-”
• SVDNet for Pedestrian Retrieval (2017, neutral): “3 MLAPG[15] - - 58.”
• Tracking the Untrackable: Learning to Track Multiple Cues with Long-Term Dependencies (2017, neutral): “MLAPG [44] 51.”
• Hi-CMD: Hierarchical Cross-Modality Disentanglement for Visible-Infrared Person Re-Identification (2019, methodology): “Feature extraction (HOG [3], LOMO [15], MLBP [16]), cross-domain matching (GSM [17]), and RGB-based person ReID (SVDNET [27], PCB [28]) methods are included for comparison.”
• RGB-Infrared Cross-Modality Person Re-identification (2017, background): “including feature learning [27, 44, 21], distance metric learning [50, 14, 20, 26, 21, 22], and end-to-end learning [19, 1, 43].”