# The Second-place Solution for CVPR 2022 SoccerNet Tracking Challenge

arXiv: https://arxiv.org/abs/2211.13481 · HTML: https://mlsanity.com/paper/2211.13481/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 13 references that are in the map.
- First posted on arXiv 24 Nov 2022.

## Builds on (most cited first, 9 of 9)

- [Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression](https://mlsanity.com/paper/1902.09630/) (2019)
- [Simple Online and Realtime Tracking with a Deep Association Metric](https://mlsanity.com/paper/1703.07402/) (2017)
- [Simple Online and Realtime Tracking](https://mlsanity.com/paper/1602.00763/) (2016)
- [ByteTrack: Multi-Object Tracking by Associating Every Detection Box](https://mlsanity.com/paper/2110.06864/) (2021)
- [Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression](https://mlsanity.com/paper/1911.08287/) (2019)
- [A Strong Baseline and Batch Normalization Neck for Deep Person Re-identification](https://mlsanity.com/paper/1906.08332/) (2019)
- [Camera Calibration and Player Localization in SoccerNet-v2 and Investigation of their Representations for Action Spotting](https://mlsanity.com/paper/2104.09333/) (2021)
- [SoccerNet 2022 Challenges Results](https://mlsanity.com/paper/2210.02365/) (2022)
- [ReMOTS: Self-Supervised Refining Multi-Object Tracking and Segmentation](https://mlsanity.com/paper/2007.03200/) (2020)

## Built on by (most cited first, 2 of 2)

- [SoccerNet Game State Reconstruction: End-to-End Athlete Tracking and Identification on a Minimap](https://mlsanity.com/paper/2404.11335/) (2024)
- [No Train Yet Gain: Towards Generic Multi-Object Tracking in Sports and Beyond](https://mlsanity.com/paper/2506.01373/) (2025)

## Abstract (arXiv)

This is our second-place solution for CVPR 2022 SoccerNet Tracking Challenge. Our method mainly includes two steps: online short-term tracking using our Cascaded Buffer-IoU (C-BIoU) Tracker, and, offline long-term tracking using appearance feature and hierarchical clustering. At each step, online tracking yielded HOTA scores near 90, and offline tracking further improved HOTA scores to around 93.2.