Deep Adaptive Feature Embedding with Local Sample Distributions for Person Re-identification

Abstract

Person re-identification (re-id) aims to match pedestrians observed by disjoint camera views. It attracts increasing attention in computer vision due to its importance to surveillance system. To combat the major challenge of cross-view visual variations, deep embedding approaches are proposed by learning a compact feature space from images such that the Euclidean distances correspond to their cross-view similarity metric. However, the global Euclidean distance cannot faithfully characterize the ideal similarity in a complex visual feature space because features of pedestrian images exhibit unknown distributions due to large variations in poses, illumination and occlusion. Moreover, intra-personal training samples within a local range are robust to guide deep embedding against uncontrolled variations, which however, cannot be captured by a global Euclidean distance. In this paper, we study the problem of person re-id by proposing a novel sampling to mine suitable \textit{positives} (i.e. intra-class) within a local range to improve the deep embedding in the context of large intra-class variations. Our method is capable of learning a deep similarity metric adaptive to local sample structure by minimizing each sample's local distances while propagating through the relationship between samples to attain the whole intra-class minimization. To this end, a novel objective function is proposed to jointly optimize similarity metric learning, local positive mining and robust deep embedding. This yields local discriminations by selecting local-ranged positive samples, and the learned features are robust to dramatic intra-class variations. Experiments on benchmarks show state-of-the-art results achieved by our method.

Published on Pattern Recognition

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Deep Learning Face Representation by Joint Identification-Verification
• Compressing Deep Convolutional Networks using Vector Quantization
• SAGA: A Fast Incremental Gradient Method With Support for Non-Strongly Convex Composite Objectives
• Learning Deep Embeddings with Histogram Loss
• Iterative Views Agreement: An Iterative Low-Rank based Structured Optimization Method to Multi-View Spectral Clustering
• Local Similarity-Aware Deep Feature Embedding
• Gated Siamese Convolutional Neural Network Architecture for Human Re-Identification
• Embedding Deep Metric for Person Re-identication A Study Against Large Variations
• A Siamese Long Short-Term Memory Architecture for Human Re-Identification
• Scalable Person Re-identification on Supervised Smoothed Manifold

Cited by in corpus (16)

• Multi-View Spectral Clustering via Structured Low-Rank Matrix Factorization
• What-and-Where to Match: Deep Spatially Multiplicative Integration Networks for Person Re-identification
• An Introduction to Person Re-identification with Generative Adversarial Networks
• Where to Focus: Deep Attention-based Spatially Recurrent Bilinear Networks for Fine-Grained Visual Recognition
• Cross-Entropy Adversarial View Adaptation for Person Re-identification
• Deep Instance-Level Hard Negative Mining Model for Histopathology Images
• Deep neural network-based classification model for Sentiment Analysis
• PAC-GAN: An Effective Pose Augmentation Scheme for Unsupervised Cross-View Person Re-identification
• Weakly Supervised Person Re-Identification
• Auto-weighted Mutli-view Sparse Reconstructive Embedding
• TPM: A GPS-based Trajectory Pattern Mining System
• Using Context Information to Enhance Simple Question Answering
• Co-regularized Multi-view Sparse Reconstruction Embedding for Dimension Reduction
• Self-Weighted Multiview Metric Learning by Maximizing the Cross Correlations
• Anomaly detecting and ranking of the cloud computing platform by multi-view learning
• Multi-feature Distance Metric Learning for Non-rigid 3D Shape Retrieval