# CFAD: Coarse-to-Fine Action Detector for Spatiotemporal Action Localization

arXiv: https://arxiv.org/abs/2008.08332 · HTML: https://mlsanity.com/paper/2008.08332/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Aug 2020.

## Built on by (most cited first, 7 of 7)

- [TubeR: Tubelet Transformer for Video Action Detection](https://mlsanity.com/paper/2104.00969/) (2021)
- [Structured Attention Composition for Temporal Action Localization](https://mlsanity.com/paper/2205.09956/) (2022)
- [A Survey on Deep Learning-based Spatio-temporal Action Detection](https://mlsanity.com/paper/2308.01618/) (2023)
- [Actor-identified Spatiotemporal Action Detection -- Detecting Who Is Doing What in Videos](https://mlsanity.com/paper/2208.12940/) (2022)
- [STMixer: A One-Stage Sparse Action Detector](https://mlsanity.com/paper/2404.09842/) (2024)
- [Towards Precise Action Spotting: Addressing Temporal Misalignment in Labels with Dynamic Label Assignment](https://mlsanity.com/paper/2504.00149/) (2025)
- [TubeLite: Lightweight Multi-Actor Spatio-Temporal Action Detection](https://mlsanity.com/paper/2607.04684/) (2026)

## Abstract (arXiv)

Most current pipelines for spatio-temporal action localization connect frame-wise or clip-wise detection results to generate action proposals, where only local information is exploited and the efficiency is hindered by dense per-frame localization. In this paper, we propose Coarse-to-Fine Action Detector (CFAD),an original end-to-end trainable framework for efficient spatio-temporal action localization. The CFAD introduces a new paradigm that first estimates coarse spatio-temporal action tubes from video streams, and then refines the tubes' location based on key timestamps. This concept is implemented by two key components, the Coarse and Refine Modules in our framework. The parameterized modeling of long temporal information in the Coarse Module helps obtain accurate initial tube estimation, while the Refine Module selectively adjusts the tube location under the guidance of key timestamps. Against other methods, theproposed CFAD achieves competitive results on action detection benchmarks of UCF101-24, UCFSports and JHMDB-21 with inference speed that is 3.3x faster than the nearest competitors.