mlsanityData to 14 Oct 2026

An Empirical Study of Vehicle Re-Identification on the AI City Challenge

arXiv 2105.09701 ↗First posted 20 May 2021cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

This paper introduces our solution for the Track2 in AI City Challenge 2021 (AICITY21). The Track2 is a vehicle re-identification (ReID) task with both the real-world data and synthetic data. We mainly focus on four points, i.e. training data, unsupervised domain-adaptive (UDA) training, post-processing, model ensembling in this challenge. (1) Both cropping training data and using synthetic data can help the model learn more discriminative features. (2) Since there is a new scenario in the test set that dose not appear in the training set, UDA methods perform well in the challenge. (3) Post-processing techniques including re-ranking, image-to-track retrieval, inter-camera fusion, etc, significantly improve final performance. (4) We ensemble CNN-based models and transformer-based models which provide different representation diversity. With aforementioned techniques, our method finally achieves 0.7445 mAP score, yielding the first place in the competition. Codes are available at https://github.com/michuanhaohao/AICITY2021_Track2_DMT.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 21 of its 33 references that are in the map. First posted on arXiv 20 May 2021.

Builds on

21 of 33 references.

References · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2016Densely Connected Convolutional Networkscited by 6,720
  4. 2020ResNeSt: Split-Attention Networkscited by 445
  5. 2021TransReID: Transformer-based Object Re-Identificationcited by 435
  6. 2019Bag of Tricks and A Strong Baseline for Deep Person Re-identificationcited by 430
  7. 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 364
  8. 2017Re-ranking Person Re-identification with k-reciprocal Encodingcited by 262
  9. 2020Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identificationcited by 215
  10. 2017Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identificationcited by 168
  11. 2019A Strong Baseline and Batch Normalization Neck for Deep Person Re-identificationcited by 155
  12. 2019CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identificationcited by 119
9 more
  1. 2019Simulating Content Consistent Vehicle Datasets with Attribute Descentcited by 36
  2. 2019A Dual-Path Model With Adaptive Attention For Vehicle Re-Identificationcited by 34
  3. 2019Vehicle Re-identification in Aerial Imagery: Dataset and Approachcited by 25
  4. 2017Learning Deep Neural Networks for Vehicle Re-ID with Visual-spatio-temporal Path Proposalscited by 23
  5. 2019Stripe-based and Attribute-aware Network: A Two-Branch Deep Model for Vehicle Re-identificationcited by 22
  6. 2020VOC-ReID: Vehicle Re-identification based on Vehicle-Orientation-Cameracited by 20
  7. 2020Multi-Domain Learning and Identity Mining for Vehicle Re-Identificationcited by 19
  8. 2018Vehicle Re-identification Using Quadruple Directional Deep Learning Featurescited by 14
  9. 20201st Place Solution to VisDA-2020: Bias Elimination for Domain Adaptive Pedestrian Re-identificationcited by 4

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 0; 2024: 3; 2025: 1.

5 papers.

  1. 2022Scalable and Real-time Multi-Camera Vehicle Detection, Re-Identification, and Trackingcited by 1
  2. 2024Enabling Cross-Camera Collaboration for Video Analytics on Distributed Smart Camerascited by 1
  3. 2024Adaptive Aspect Ratios with Patch-Mixup-ViT-based Vehicle ReIDcited by 1
  4. 2025Exploring the Camera Bias of Person Re-identificationcited by 1
  5. 2024Study on Aspect Ratio Variability toward Robustness of Vision Transformer-based Vehicle Re-identificationcited by 0