A Strong Baseline and Batch Normalization Neck for Deep Person Re-identification

Abstract

This study explores a simple but strong baseline for person re-identification (ReID). Person ReID with deep neural networks has progressed and achieved high performance in recent years. However, many state-of-the-art methods design complex network structures and concatenate multi-branch features. In the literature, some effective training tricks briefly appear in several papers or source codes. The present study collects and evaluates these effective training tricks in person ReID. By combining these tricks, the model achieves 94.5% rank-1 and 85.9% mean average precision on Market1501 with only using the global features of ResNet50. The performance surpasses all existing global- and part-based baselines in person ReID. We propose a novel neck structure named as batch normalization neck (BNNeck). BNNeck adds a batch normalization layer after global pooling layer to separate metric and classification losses into two different feature spaces because we observe they are inconsistent in one embedding space. Extended experiments show that BNNeck can boost the baseline, and our baseline can improve the performance of existing state-of-the-art methods. Our codes and models are available at: https://github.com/michuanhaohao/reid-strong-baseline.

arXiv commentAccepted by IEEE Transactions on Multimedia. This is the submitted journal version of the oral paper [arXiv:1903.07071] in CVPRW'19. Code are avaliable at: https://github.com/michuanhaohao/reid-strong-baseline

References in corpus (5)

• FaceNet: A Unified Embedding for Face Recognition and Clustering
• A Discriminatively Learned CNN Embedding for Person Re-identification
• End-to-End Comparative Attention Networks for Person Re-identification
• GLAD: Global-Local-Alignment Descriptor for Pedestrian Retrieval
• SphereReID: Deep Hypersphere Manifold Embedding for Person Re-Identification

Cited by in corpus (16)

• Parameter Sharing Exploration and Hetero-Center based Triplet Loss for Visible-Thermal Person Re-Identification
• Exploring Spatial Significance via Hybrid Pyramidal Graph Network for Vehicle Re-identification
• CMTR: Cross-modality Transformer for Visible-infrared Person Re-identification
• Learning Multi-Attention Context Graph for Group-Based Re-Identification
• Strong but Simple Baseline with Dual-Granularity Triplet Loss for Visible-Thermal Person Re-Identification
• Leaning Compact and Representative Features for Cross-Modality Person Re-Identification
• Style Alignment based Dynamic Observation Method for UAV-View Geo-localization
• Hierarchical Bi-Directional Feature Perception Network for Person Re-Identification
• Leveraging Ensembles and Self-Supervised Learning for Fully-Unsupervised Person Re-Identification and Text Authorship Attribution
• Enhancing Visible-Infrared Person Re-identification with Modality- and Instance-aware Visual Prompt Learning
• Dual-Refinement: Joint Label and Feature Refinement for Unsupervised Domain Adaptive Person Re-Identification
• FeatureSORT: Essential Features for Effective Tracking
• CORE-ReID V2: Advancing the Domain Adaptation for Object Re-Identification with Optimized Training and Ensemble Fusion
• A Strong Baseline for Fashion Retrieval with Person Re-Identification Models
• Text-Guided Multi-Instance Learning for Scoliosis Screening via Gait Video Analysis
• Lifelong Person Search