# TrackNetV4: Enhancing Fast Sports Object Tracking with Motion Attention Maps

arXiv: https://arxiv.org/abs/2409.14543 · HTML: https://mlsanity.com/paper/2409.14543/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 22 Sep 2024.

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

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors](https://mlsanity.com/paper/2207.02696/) (2022)
- [TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications](https://mlsanity.com/paper/1907.03698/) (2019)
- [Cross-modal Prompts: Adapting Large Pre-trained Models for Audio-Visual Downstream Tasks](https://mlsanity.com/paper/2311.05152/) (2023)
- [FinePOSE: Fine-Grained Prompt-Driven 3D Human Pose Estimation via Diffusion Models](https://mlsanity.com/paper/2405.05216/) (2024)
- [High-order Tensor Pooling with Attention for Action Recognition](https://mlsanity.com/paper/2110.05216/) (2021)
- [Flow Dynamics Correction for Action Recognition](https://mlsanity.com/paper/2310.10059/) (2023)
- [Taylor Videos for Action Recognition](https://mlsanity.com/paper/2402.03019/) (2024)
- [Motion meets Attention: Video Motion Prompts](https://mlsanity.com/paper/2407.03179/) (2024)

## Built on by (most cited first, 11 of 11)

- [Quo Vadis, Anomaly Detection? LLMs and VLMs in the Spotlight](https://mlsanity.com/paper/2412.18298/) (2024)
- [TT3D: Table Tennis 3D Reconstruction](https://mlsanity.com/paper/2504.10035/) (2025)
- [Learning Time in Static Classifiers](https://mlsanity.com/paper/2511.12321/) (2025)
- [Evolving Skeletons: Motion Dynamics in Action Recognition](https://mlsanity.com/paper/2501.02593/) (2025)
- [BlurBall: Joint Ball and Motion Blur Estimation for Table Tennis Ball Tracking](https://mlsanity.com/paper/2509.18387/) (2025)
- [TT4D: A Pipeline and Dataset for Table Tennis 4D Reconstruction From Monocular Videos](https://mlsanity.com/paper/2605.01234/) (2026)
- [End-to-End Unmixing with Material Prompts for Hyperspectral Object Tracking](https://mlsanity.com/paper/2605.20569/) (2026)
- [TrackNetV5: Residual-Driven Spatio-Temporal Refinement and Motion Direction Decoupling for Fast Object Tracking](https://mlsanity.com/paper/2512.02789/) (2025)
- [One-Shot Badminton Shuttle Detection for Mobile Robots](https://mlsanity.com/paper/2603.06691/) (2026)
- [Rule-Constrained Assignment for Cue-Ball Identification in Broadcast Snooker](https://mlsanity.com/paper/2609.23450/) (2026)
- [BMASH: Ball-Motion-Aware Soccer Header Spotting](https://mlsanity.com/paper/2609.39300/) (2026)

## Abstract (arXiv)

Accurately detecting and tracking high-speed, small objects, such as balls in sports videos, is challenging due to factors like motion blur and occlusion. Although recent deep learning frameworks like TrackNetV1, V2, and V3 have advanced tennis ball and shuttlecock tracking, they often struggle in scenarios with partial occlusion or low visibility. This is primarily because these models rely heavily on visual features without explicitly incorporating motion information, which is crucial for precise tracking and trajectory prediction. In this paper, we introduce an enhancement to the TrackNet family by fusing high-level visual features with learnable motion attention maps through a motion-aware fusion mechanism, effectively emphasizing the moving ball's location and improving tracking performance. Our approach leverages frame differencing maps, modulated by a motion prompt layer, to highlight key motion regions over time. Experimental results on the tennis ball and shuttlecock datasets show that our method enhances the tracking performance of both TrackNetV2 and V3. We refer to our lightweight, plug-and-play solution, built on top of the existing TrackNet, as TrackNetV4.