Features for Multi-target Multi-camera Tracking and Re-identification

By Ergys Ristani, Carlo Tomasi. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018

554 citations recorded by the source corpus.

Multi-Target Multi-Camera Tracking (MTMCT) tracks many people through video taken from several cameras. Person Re-Identification (Re-ID) retrieves from a gallery images of people similar to a person query image. We learn good features for both MTMCT and Re-ID with a convolutional neural network. Our contributions include an adaptive weighted triplet loss for training and a new technique for hard-identity mining. Our method outperforms the state of the art both on the DukeMTMC benchmarks for tracking, and on the Market-1501 and DukeMTMC-ReID benchmarks for Re-ID. We examine the correlation between good Re-ID and good MTMCT scores, and perform ablation studies to elucidate the contributions of the main components of our system. Code is available1.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/c0f01b8174a632448c20eb5472cd9d5b2c595e39

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yi Yang
2. Dacheng Tao
3. Zhedong Zheng
4. Liang Zheng
5. Tim Meinhardt
6. Changxing Ding
7. L. Leal-Taix'e
8. Jan Kautz

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Luc Van Gool
3. Yi Yang
4. Ming-Hsuan Yang
5. Xiaogang Wang
6. Liang Zheng
7. Alan L. Yuille
8. Trevor Darrell

Citations in the corpus

• Bag of Tricks and a Strong Baseline for Deep Person Re-Identification (2019, methodology): “Throughout all methods that only use global features, our strong baseline beats AWTL[12] by more than 10% mAP on both Market1501 and DukeMTMC-reID.”
• Tracking Without Bells and Whistles (2019, methodology): “Recently, [51] showed the importance of learned reID features for multi-object tracking.”
• Batch Coherence-Driven Network for Part-Aware Person Re-Identification (2020, background): “, video surveillance [1], searching for missing children, etc.”
• Learning a Neural Solver for Multiple Object Tracking (2019, background): “In [56], authors showed the importance of learned reID features for multi-object tracking.”
• Joint Discriminative and Generative Learning for Person Re-Identification (2019, methodology): “Some methods combine identification loss with verification loss [48,55], others apply triplet loss with hard sample mining [6, 13, 33].”
• ABD-Net: Attentive but Diverse Person Re-Identification (2019, unknown): “51 AWTL(2-stream) [57] (2018 CVPR) 79.”
• TrackFormer: Multi-Object Tracking with Transformers (2021, unknown): “Appearance driven methods capitalize on increasingly powerful image recognition backbones to track objects by relying on similarity measures given by twin neural networks [23], learned reID features [32, 41], detection candidate selection [8] or affinity estimation [10].”
• Weakly Supervised Complementary Parts Models for Fine-Grained Image Classification From the Bottom Up (2019, background): “This is a common pipeline in many high-level vision tasks, such as object detection [10, 14, 16], tracking [43, 33, 38], and scene under-”