Semantic Consistency and Identity Mapping Multi-Component Generative Adversarial Network for Person Re-Identification
Not on the map. Why?
Abstract (arXiv)
In a real world environment, person re-identification (Re-ID) is a challenging task due to variations in lighting conditions, viewing angles, pose and occlusions. Despite recent performance gains, current person Re-ID algorithms still suffer heavily when encountering these variations. To address this problem, we propose a semantic consistency and identity mapping multi-component generative adversarial network (SC-IMGAN) which provides style adaptation from one to many domains. To ensure that transformed images are as realistic as possible, we propose novel identity mapping and semantic consistency losses to maintain identity across the diverse domains. For the Re-ID task, we propose a joint verification-identification quartet network which is trained with generated and real images, followed by an effective quartet loss for verification. Our proposed method outperforms state-of-the-art techniques on six challenging person Re-ID datasets: CUHK01, CUHK03, VIPeR, PRID2011, iLIDS and Market-1501.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 31 of its 52 references that are in the map. First posted on arXiv 28 Apr 2021.
Builds on
31 of 52 references.
References · 31
- 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
- 2017StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translationcited by 1,039
- 2017Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networkscited by 707
- 2017Person Transfer GAN to Bridge Domain Gap for Person Re-Identificationcited by 480
- 2017DualGAN: Unsupervised Dual Learning for Image-to-Image Translationcited by 310
- 2017Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitrocited by 281
- 2017Re-ranking Person Re-identification with k-reciprocal Encodingcited by 262
- 2016Coupled Generative Adversarial Networkscited by 216
- 2017Beyond triplet loss: a deep quadruplet network for person re-identificationcited by 172
- 2016Unsupervised Cross-Domain Image Generationcited by 125
- 2016A Discriminatively Learned CNN Embedding for Person Re-identificationcited by 78
- 2017Unsupervised Person Re-identification: Clustering and Fine-tuningcited by 73
19 more
- 2017Camera Style Adaptation for Person Re-identificationcited by 68
- 2018Multi-Level Factorisation Net for Person Re-Identificationcited by 58
- 2017Learning Deep Context-aware Features over Body and Latent Parts for Person Re-identificationcited by 52
- 2018Attention-Aware Compositional Network for Person Re-identificationcited by 50
- 2018Domain Adaptation through Synthesis for Unsupervised Person Re-identificationcited by 44
- 2016End-to-End Comparative Attention Networks for Person Re-identificationcited by 42
- 2016Learning Deep Feature Representations with Domain Guided Dropout for Person Re-identificationcited by 41
- 2018Multi-component Image Translation for Deep Domain Generalizationcited by 40
- 2016Gated Siamese Convolutional Neural Network Architecture for Human Re-Identificationcited by 34
- 2015Deep Feature Learning with Relative Distance Comparison for Person Re-identificationcited by 25
- 2016Pixel-Level Domain Transfercited by 25
- 2017Scalable Person Re-identification on Supervised Smoothed Manifoldcited by 24
- 2019Unsupervised Person Re-identification by Deep Asymmetric Metric Embeddingcited by 22
- 2018Resource Aware Person Re-identification across Multiple Resolutionscited by 21
- 2016Embedding Deep Metric for Person Re-identication A Study Against Large Variationscited by 16
- 2015Deep Ranking for Person Re-identification via Joint Representation Learningcited by 10
- 2016PersonNet: Person Re-identification with Deep Convolutional Neural Networkscited by 9
- 2016A Multi-task Deep Network for Person Re-identificationcited by 9
- 2018A Deep Four-Stream Siamese Convolutional Neural Network with Joint Verification and Identification Loss for Person Re-detectioncited by 4
Built on by
Citing papers per year: 2021: 1.
1 paper.