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

arXiv: https://arxiv.org/abs/2105.09701 · HTML: https://mlsanity.com/paper/2105.09701/

## 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 (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [ResNeSt: Split-Attention Networks](https://mlsanity.com/paper/2004.08955/) (2020)
- [TransReID: Transformer-based Object Re-Identification](https://mlsanity.com/paper/2102.04378/) (2021)
- [Bag of Tricks and A Strong Baseline for Deep Person Re-identification](https://mlsanity.com/paper/1903.07071/) (2019)
- [Two at Once: Enhancing Learning and Generalization Capacities via IBN-Net](https://mlsanity.com/paper/1807.09441/) (2018)
- [Re-ranking Person Re-identification with k-reciprocal Encoding](https://mlsanity.com/paper/1701.08398/) (2017)
- [Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification](https://mlsanity.com/paper/2001.01526/) (2020)
- [Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification](https://mlsanity.com/paper/1711.07027/) (2017)
- [A Strong Baseline and Batch Normalization Neck for Deep Person Re-identification](https://mlsanity.com/paper/1906.08332/) (2019)
- [CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification](https://mlsanity.com/paper/1903.09254/) (2019)
- [Simulating Content Consistent Vehicle Datasets with Attribute Descent](https://mlsanity.com/paper/1912.08855/) (2019)
- [A Dual-Path Model With Adaptive Attention For Vehicle Re-Identification](https://mlsanity.com/paper/1905.03397/) (2019)
- [Vehicle Re-identification in Aerial Imagery: Dataset and Approach](https://mlsanity.com/paper/1904.01400/) (2019)
- [Learning Deep Neural Networks for Vehicle Re-ID with Visual-spatio-temporal Path Proposals](https://mlsanity.com/paper/1708.03918/) (2017)
- [Stripe-based and Attribute-aware Network: A Two-Branch Deep Model for Vehicle Re-identification](https://mlsanity.com/paper/1910.05549/) (2019)
- [VOC-ReID: Vehicle Re-identification based on Vehicle-Orientation-Camera](https://mlsanity.com/paper/2004.09164/) (2020)
- [Multi-Domain Learning and Identity Mining for Vehicle Re-Identification](https://mlsanity.com/paper/2004.10547/) (2020)
- [Vehicle Re-identification Using Quadruple Directional Deep Learning Features](https://mlsanity.com/paper/1811.05163/) (2018)
- [1st Place Solution to VisDA-2020: Bias Elimination for Domain Adaptive Pedestrian Re-identification](https://mlsanity.com/paper/2012.13498/) (2020)

## Built on by (most cited first, 5 of 5)

- [Scalable and Real-time Multi-Camera Vehicle Detection, Re-Identification, and Tracking](https://mlsanity.com/paper/2204.07442/) (2022)
- [Enabling Cross-Camera Collaboration for Video Analytics on Distributed Smart Cameras](https://mlsanity.com/paper/2401.14132/) (2024)
- [Adaptive Aspect Ratios with Patch-Mixup-ViT-based Vehicle ReID](https://mlsanity.com/paper/2411.06297/) (2024)
- [Exploring the Camera Bias of Person Re-identification](https://mlsanity.com/paper/2502.10195/) (2025)
- [Study on Aspect Ratio Variability toward Robustness of Vision Transformer-based Vehicle Re-identification](https://mlsanity.com/paper/2407.07842/) (2024)

## 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.