Parameter-Free Spatial Attention Network for Person Re-Identification

Abstract

Global average pooling (GAP) allows to localize discriminative information for recognition [40]. While GAP helps the convolution neural network to attend to the most discriminative features of an object, it may suffer if that information is missing e.g. due to camera viewpoint changes. To circumvent this issue, we argue that it is advantageous to attend to the global configuration of the object by modeling spatial relations among high-level features. We propose a novel architecture for Person Re-Identification, based on a novel parameter-free spatial attention layer introducing spatial relations among the feature map activations back to the model. Our spatial attention layer consistently improves the performance over the model without it. Results on four benchmarks demonstrate a superiority of our model over the state-of-the-art achieving rank-1 accuracy of 94.7% on Market-1501, 89.0% on DukeMTMC-ReID, 74.9% on CUHK03-labeled and 69.7% on CUHK03-detected.

References in corpus (8)

• Random Erasing Data Augmentation
• Learning Deep Features for Discriminative Localization
• Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking
• Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)
• Attention-Aware Compositional Network for Person Re-identification
• Dual Attention Matching Network for Context-Aware Feature Sequence based Person Re-Identification
• Constrained Deep Metric Learning for Person Re-identification
• Automatic Discovery and Optimization of Parts for Image Classification

Cited by in corpus (10)

• NL-LinkNet: Toward Lighter but More Accurate Road Extraction with Non-Local Operations
• FPB: Feature Pyramid Branch for Person Re-Identification
• Towards Practical Implementations of Person Re-Identification from Full Video Frames
• Robust Person Re-Identification through Contextual Mutual Boosting
• Incremental Embedding Learning via Zero-Shot Translation
• Calibrated Domain-Invariant Learning for Highly Generalizable Large Scale Re-Identification
• Iterative and Adaptive Sampling with Spatial Attention for Black-Box Model Explanations
• Interflow: Aggregating Multi-layer Feature Mappings with Attention Mechanism
• Beyond Universal Person Re-ID Attack
• Pose Invariant Person Re-Identification using Robust Pose-transformation GAN