SportMamba: Adaptive Non-Linear Multi-Object Tracking with State Space Models for Team Sports
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916777595764736 |
|---|---|
| author | Khanna, Dheeraj Bright, Jerrin Chen, Yuhao Zelek, John S. |
| author_facet | Khanna, Dheeraj Bright, Jerrin Chen, Yuhao Zelek, John S. |
| contents | Multi-object tracking (MOT) in team sports is particularly challenging due to the fast-paced motion and frequent occlusions resulting in motion blur and identity switches, respectively. Predicting player positions in such scenarios is particularly difficult due to the observed highly non-linear motion patterns. Current methods are heavily reliant on object detection and appearance-based tracking, which struggle to perform in complex team sports scenarios, where appearance cues are ambiguous and motion patterns do not necessarily follow a linear pattern. To address these challenges, we introduce SportMamba, an adaptive hybrid MOT technique specifically designed for tracking in dynamic team sports. The technical contribution of SportMamba is twofold. First, we introduce a mamba-attention mechanism that models non-linear motion by implicitly focusing on relevant embedding dependencies. Second, we propose a height-adaptive spatial association metric to reduce ID switches caused by partial occlusions by accounting for scale variations due to depth changes. Additionally, we extend the detection search space with adaptive buffers to improve associations in fast-motion scenarios. Our proposed technique, SportMamba, demonstrates state-of-the-art performance on various metrics in the SportsMOT dataset, which is characterized by complex motion and severe occlusion. Furthermore, we demonstrate its generalization capability through zero-shot transfer to VIP-HTD, an ice hockey dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_03335 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SportMamba: Adaptive Non-Linear Multi-Object Tracking with State Space Models for Team Sports Khanna, Dheeraj Bright, Jerrin Chen, Yuhao Zelek, John S. Computer Vision and Pattern Recognition Multi-object tracking (MOT) in team sports is particularly challenging due to the fast-paced motion and frequent occlusions resulting in motion blur and identity switches, respectively. Predicting player positions in such scenarios is particularly difficult due to the observed highly non-linear motion patterns. Current methods are heavily reliant on object detection and appearance-based tracking, which struggle to perform in complex team sports scenarios, where appearance cues are ambiguous and motion patterns do not necessarily follow a linear pattern. To address these challenges, we introduce SportMamba, an adaptive hybrid MOT technique specifically designed for tracking in dynamic team sports. The technical contribution of SportMamba is twofold. First, we introduce a mamba-attention mechanism that models non-linear motion by implicitly focusing on relevant embedding dependencies. Second, we propose a height-adaptive spatial association metric to reduce ID switches caused by partial occlusions by accounting for scale variations due to depth changes. Additionally, we extend the detection search space with adaptive buffers to improve associations in fast-motion scenarios. Our proposed technique, SportMamba, demonstrates state-of-the-art performance on various metrics in the SportsMOT dataset, which is characterized by complex motion and severe occlusion. Furthermore, we demonstrate its generalization capability through zero-shot transfer to VIP-HTD, an ice hockey dataset. |
| title | SportMamba: Adaptive Non-Linear Multi-Object Tracking with State Space Models for Team Sports |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.03335 |