SVDNet for Pedestrian Retrieval

Abstract

This paper proposes the SVDNet for retrieval problems, with focus on the application of person re-identification (re-ID). We view each weight vector within a fully connected (FC) layer in a convolutional neuron network (CNN) as a projection basis. It is observed that the weight vectors are usually highly correlated. This problem leads to correlations among entries of the FC descriptor, and compromises the retrieval performance based on the Euclidean distance. To address the problem, this paper proposes to optimize the deep representation learning process with Singular Vector Decomposition (SVD). Specifically, with the restraint and relaxation iteration (RRI) training scheme, we are able to iteratively integrate the orthogonality constraint in CNN training, yielding the so-called SVDNet. We conduct experiments on the Market-1501, CUHK03, and Duke datasets, and show that RRI effectively reduces the correlation among the projection vectors, produces more discriminative FC descriptors, and significantly improves the re-ID accuracy. On the Market-1501 dataset, for instance, rank-1 accuracy is improved from 55.3% to 80.5% for CaffeNet, and from 73.8% to 82.3% for ResNet-50.

accepted as spotlight to ICCV 2017

References in corpus (9)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Going Deeper with Convolutions
• Person Re-identification: Past, Present and Future
• Pedestrian Alignment Network for Large-scale Person Re-identification
• Deep Transfer Learning for Person Re-identification
• Looking Beyond Appearances: Synthetic Training Data for Deep CNNs in Re-identification
• Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro
• Pose Invariant Embedding for Deep Person Re-identification
• All You Need is Beyond a Good Init: Exploring Better Solution for Training Extremely Deep Convolutional Neural Networks with Orthonormality and Modulation

Cited by in corpus (58)

• Random Erasing Data Augmentation
• SphereReID: Deep Hypersphere Manifold Embedding for Person Re-Identification
• Harmonious Attention Network for Person Re-Identification
• Omni-Scale Feature Learning for Person Re-Identification
• Multi-task Learning with Coarse Priors for Robust Part-aware Person Re-identification
• Re-ID done right: towards good practices for person re-identification
• Attention-Aware Compositional Network for Person Re-identification
• The Devil is in the Middle: Exploiting Mid-level Representations for Cross-Domain Instance Matching
• Dual Attention Matching Network for Context-Aware Feature Sequence based Person Re-Identification
• Improved Person Re-Identification Based on Saliency and Semantic Parsing with Deep Neural Network Models
• Let Features Decide for Themselves: Feature Mask Network for Person Re-identification
• Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification
• Spatially and Temporally Efficient Non-local Attention Network for Video-based Person Re-Identification
• Camera Style Adaptation for Person Re-identification
• Learning Shape Representations for Clothing Variations in Person Re-Identification
• Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training
• ES-Net: Erasing Salient Parts to Learn More in Re-Identification
• Pose-Normalized Image Generation for Person Re-identification
• Deep Association Learning for Unsupervised Video Person Re-identification
• Towards Good Practices on Building Effective CNN Baseline Model for Person Re-identification
• CA3Net: Contextual-Attentional Attribute-Appearance Network for Person Re-Identification
• Joint Disentangling and Adaptation for Cross-Domain Person Re-Identification
• Person Re-identification with Deep Similarity-Guided Graph Neural Network
• Occluded Person Re-identification
• Spectral Feature Transformation for Person Re-identification
• Spatial-Temporal Person Re-identification
• Adaptively Connected Neural Networks
• Query Attack via Opposite-Direction Feature:Towards Robust Image Retrieval
• Deep Co-attention based Comparators For Relative Representation Learning in Person Re-identification
• Improving Deep Visual Representation for Person Re-identification by Global and Local Image-language Association
• An Evaluation of Deep CNN Baselines for Scene-Independent Person Re-Identification
• Adaptation and Re-Identification Network: An Unsupervised Deep Transfer Learning Approach to Person Re-Identification
• End-to-End Deep Kronecker-Product Matching for Person Re-identification
• Weighted Bilinear Coding over Salient Body Parts for Person Re-identification
• Where-and-When to Look: Deep Siamese Attention Networks for Video-based Person Re-identification
• Backbone Can Not be Trained at Once: Rolling Back to Pre-trained Network for Person Re-Identification
• Collaborative Attention Network for Person Re-identification
• HorNet: A Hierarchical Offshoot Recurrent Network for Improving Person Re-ID via Image Captioning
• Adaptive Re-ranking of Deep Feature for Person Re-identification
• SuperFront: From Low-resolution to High-resolution Frontal Face Synthesis
• See More, Know More: Unsupervised Video Object Segmentation with Co-Attention Siamese Networks
• Self Attention Grid for Person Re-Identification
• Vehicle Re-Identification in Context
• Hierarchical and Efficient Learning for Person Re-Identification
• Homocentric Hypersphere Feature Embedding for Person Re-identification
• Operator-in-the-Loop Deep Sequential Multi-camera Feature Fusion for Person Re-identification
• Progressive Multi-stage Feature Mix for Person Re-Identification
• Deep Person Re-Identification with Improved Embedding and Efficient Training
• SCPNet: Spatial-Channel Parallelism Network for Joint Holistic and Partial Person Re-Identification
• Adversarial Binary Coding for Efficient Person Re-identification
• A framework with updateable joint images re-ranking for Person Re-identification
• Imbalance Robust Softmax for Deep Embeeding Learning
• Dynamic Metric Learning: Towards a Scalable Metric Space to Accommodate Multiple Semantic Scales
• Compact Deep Aggregation for Set Retrieval
• VMRFANet:View-Specific Multi-Receptive Field Attention Network for Person Re-identification
• Person Re-identification with Bias-controlled Adversarial Training
• Pseudo-positive regularization for deep person re-identification
• Cross-Resolution Person Re-identification with Deep Antithetical Learning