# Decoupling Spatio-Temporal Adapter for Fine-Grained Badminton Action Localization

arXiv: https://arxiv.org/abs/2605.23355 · HTML: https://mlsanity.com/paper/2605.23355/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 40 references that are in the map.
- First posted on arXiv 22 May 2026.

## Builds on (most cited first, 26 of 26)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://mlsanity.com/paper/2203.12602/) (2022)
- [A Closer Look at Spatiotemporal Convolutions for Action Recognition](https://mlsanity.com/paper/1711.11248/) (2017)
- [BMN: Boundary-Matching Network for Temporal Action Proposal Generation](https://mlsanity.com/paper/1907.09702/) (2019)
- [ActionFormer: Localizing Moments of Actions with Transformers](https://mlsanity.com/paper/2202.07925/) (2022)
- [BSN: Boundary Sensitive Network for Temporal Action Proposal Generation](https://mlsanity.com/paper/1806.02964/) (2018)
- [ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning](https://mlsanity.com/paper/2206.13559/) (2022)
- [Frozen CLIP Models are Efficient Video Learners](https://mlsanity.com/paper/2208.03550/) (2022)
- [End-to-end Temporal Action Detection with Transformer](https://mlsanity.com/paper/2106.10271/) (2021)
- [AIM: Adapting Image Models for Efficient Video Action Recognition](https://mlsanity.com/paper/2302.03024/) (2023)
- [Relaxed Transformer Decoders for Direct Action Proposal Generation](https://mlsanity.com/paper/2102.01894/) (2021)
- [TALLFormer: Temporal Action Localization with a Long-memory Transformer](https://mlsanity.com/paper/2204.01680/) (2022)
- [ReAct: Temporal Action Detection with Relational Queries](https://mlsanity.com/paper/2207.07097/) (2022)
- [End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames](https://mlsanity.com/paper/2311.17241/) (2023)
- [MonoTrack: Shuttle trajectory reconstruction from monocular badminton video](https://mlsanity.com/paper/2204.01899/) (2022)
- [ShuttleNet: Position-aware Fusion of Rally Progress and Player Styles for Stroke Forecasting in Badminton](https://mlsanity.com/paper/2112.01044/) (2021)
- [Towards Structured Analysis of Broadcast Badminton Videos](https://mlsanity.com/paper/1712.08714/) (2017)
- [ShuttleSet: A Human-Annotated Stroke-Level Singles Dataset for Badminton Tactical Analysis](https://mlsanity.com/paper/2306.04948/) (2023)
- [ETAD: Training Action Detection End to End on a Laptop](https://mlsanity.com/paper/2205.07134/) (2022)
- [Estimation of control area in badminton doubles with pose information from top and back view drone videos](https://mlsanity.com/paper/2305.04247/) (2023)
- [BADGE: BADminton report Generation and Evaluation with LLM](https://mlsanity.com/paper/2406.18116/) (2024)
- [VIRD: Immersive Match Video Analysis for High-Performance Badminton Coaching](https://mlsanity.com/paper/2307.12539/) (2023)
- [Automated Hit-frame Detection for Badminton Match Analysis](https://mlsanity.com/paper/2307.16000/) (2023)
- [Counterfactual Explanation-Based Badminton Motion Guidance Generation Using Wearable Sensors](https://mlsanity.com/paper/2405.11802/) (2024)

## Abstract (arXiv)

Temporal Action Localization (TAL) has been extensively studied in generic video understanding, while fine-grained sports scenarios, such as professional badminton, remain underexplored due to their complex and subtle spatio-temporal dynamics. In this paper, we focus on fine-grained TAL in professional badminton videos and introduce a new benchmark dataset, Fine-Badminton, which consists of 31 matches with 29 fine-grained stroke categories, covering 2104 rallies and 27597 annotated actions. To effectively capture the intricate motion patterns in such scenarios, we propose a Decoupling Spatio-Temporal Adapter (DSTA), which enables efficient modeling of spatio-temporal features within a parameter-efficient framework. Specifically, DSTA decomposes motion representation into three parallel branches, capturing temporal dynamics as well as vertical and horizontal spatial variations. The design allows the model to better distinguish subtle differences among fine-grained actions. Extensive experiments on both the Fine-Badminton dataset and the ShuttleSet benchmark demonstrate that the proposed method achieves state-of-the-art performance while introducing only a marginal increase in computational and parameter cost. These results validate the effectiveness and efficiency of the proposed approach for fine-grained temporal action localization.