Deep Representation Learning with Part Loss for Person Re-Identification
arXiv:1707.00798 · doi:10.1109/TIP.2019.2891888
Abstract
Learning discriminative representations for unseen person images is critical for person Re-Identification (ReID). Most of current approaches learn deep representations in classification tasks, which essentially minimize the empirical classification risk on the training set. As shown in our experiments, such representations commonly focus on several body parts discriminative to the training set, rather than the entire human body. Inspired by the structural risk minimization principle in SVM, we revise the traditional deep representation learning procedure to minimize both the empirical classification risk and the representation learning risk. The representation learning risk is evaluated by the proposed part loss, which automatically generates several parts for an image, and computes the person classification loss on each part separately. Compared with traditional global classification loss, simultaneously considering multiple part loss enforces the deep network to focus on the entire human body and learn discriminative representations for different parts. Experimental results on three datasets, i.e., Market1501, CUHK03, VIPeR, show that our representation outperforms the existing deep representations.
9 pages, 9 figures
References in corpus (7)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Caffe: Convolutional Architecture for Fast Feature Embedding
- Going Deeper with Convolutions
- GLAD: Global-Local-Alignment Descriptor for Pedestrian Retrieval
- Pose Invariant Embedding for Deep Person Re-identification
- Deep Metric Learning for Practical Person Re-Identification
Cited by in corpus (74)
- AlignedReID: Surpassing Human-Level Performance in Person Re-Identification
- A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localization
- SphereReID: Deep Hypersphere Manifold Embedding for Person Re-Identification
- An overview of artificial intelligence techniques for diagnosis of Schizophrenia based on magnetic resonance imaging modalities: Methods, challenges, and future works
- Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)
- Person Re-identification via Attention Pyramid
- Second-order Non-local Attention Networks for Person Re-identification
- Multi-task Learning with Coarse Priors for Robust Part-aware Person Re-identification
- FragNet: Writer Identification using Deep Fragment Networks
- TransReID: Transformer-based Object Re-Identification
- Incomplete Descriptor Mining with Elastic Loss for Person Re-Identification
- Semantically Self-Aligned Network for Text-to-Image Part-aware Person Re-identification
- ABD-Net: Attentive but Diverse Person Re-Identification
- Deep-Person: Learning Discriminative Deep Features for Person Re-Identification
- Horizontal Pyramid Matching for Person Re-identification
- Global-Local Temporal Representations For Video Person Re-Identification
- Adaptive Attribute and Structure Subspace Clustering Network
- Joint Person Objectness and Repulsion for Person Search
- Pose-driven Deep Convolutional Model for Person Re-identification
- Exploiting Robust Unsupervised Video Person Re-identification
- CDPM: Convolutional Deformable Part Models for Semantically Aligned Person Re-identification
- Learning Shape Representations for Clothing Variations in Person Re-Identification
- Dynamic Dual-Attentive Aggregation Learning for Visible-Infrared Person Re-Identification
- Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training
- Batch Coherence-Driven Network for Part-aware Person Re-Identification
- Receptive Multi-granularity Representation for Person Re-Identification
- Pose-Normalized Image Generation for Person Re-identification
- AANet: Attribute Attention Network for Person Re-Identifications
- Batch DropBlock Network for Person Re-identification and Beyond
- Multilinear subspace learning for person re-identification based fusion of high order tensor features
- Beyond Triplet Loss: Person Re-identification with Fine-grained Difference-aware Pairwise Loss
- OH-Former: Omni-Relational High-Order Transformer for Person Re-Identification
- Densely Semantically Aligned Person Re-Identification
- Hierarchical Bi-Directional Feature Perception Network for Person Re-Identification
- Cross-Spectrum Dual-Subspace Pairing for RGB-infrared Cross-Modality Person Re-Identification
- A heterogeneous branch and multi-level classification network for person re-identification
- Improving Description-based Person Re-identification by Multi-granularity Image-text Alignments
- Semantics-Aligned Representation Learning for Person Re-identification
- Soft-ranking Label Encoding for Robust Facial Age Estimation
- Grafted network for person re-identification
- An Evaluation of Deep CNN Baselines for Scene-Independent Person Re-Identification
- Multi-Scale Body-Part Mask Guided Attention for Person Re-identification
- Weighted Bilinear Coding over Salient Body Parts for Person Re-identification
- Going Deeper into Semi-supervised Person Re-identification
- LVreID: Person Re-Identification with Long Sequence Videos
- Learning Longterm Representations for Person Re-Identification Using Radio Signals
- Discovering Underlying Person Structure Pattern with Relative Local Distance for Person Re-identification
- Exploring Modality-shared Appearance Features and Modality-invariant Relation Features for Cross-modality Person Re-Identification
- Self-Supervised Graph Representation Learning via Global Context Prediction
- DotSCN: Group Re-identification via Domain-Transferred Single and Couple Representation Learning
- Pedestrian re-identification based on Tree branch network with local and global learning
- TAL: Two-stream Adaptive Learning for Generalizable Person Re-identification
- Collaborative Attention Network for Person Re-identification
- Diversity-Achieving Slow-DropBlock Network for Person Re-Identification
- In Defense of the Classification Loss for Person Re-Identification
- A Convolutional Baseline for Person Re-Identification Using Vision and Language Descriptions
- Relation Network for Person Re-identification
- Improved Res2Net model for Person re-identification
- Unsupervised Domain Adaptive Learning via Synthetic Data for Person Re-identification
- Multigranular Visual-Semantic Embedding for Cloth-Changing Person Re-identification
- Person image generation with semantic attention network for person re-identification
- Learning Deep Representations by Mutual Information for Person Re-identification
- Learning Portrait Style Representations
- Take More Positives: An Empirical Study of Contrastive Learing in Unsupervised Person Re-Identification
- Micro-supervised Disturbance Learning: A Perspective of Representation Probability Distribution
- Progressive Learning Algorithm for Efficient Person Re-Identification
- Person Re-Identification using Deep Learning Networks: A Systematic Review
- Perceive Where to Focus: Learning Visibility-aware Part-level Features for Partial Person Re-identification
- SCPNet: Spatial-Channel Parallelism Network for Joint Holistic and Partial Person Re-Identification
- Attention Driven Person Re-identification
- Learning Incremental Triplet Margin for Person Re-identification
- Beyond Triplet Loss: Meta Prototypical N-tuple Loss for Person Re-identification
- Unified Multifaceted Feature Learning for Person Re-Identification
- Local-Global Associative Frame Assemble in Video Re-ID