FineSports: A Multi-Person Hierarchical Sports Video Dataset for Fine-Grained Action Understanding
Authors:
- 16 June 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 21773-21782
Abstract
Fine-grained action analysis in multi-person sports is complex due to athletes' quick movements and intense physical confrontations, which result in severe visual obstructions in most scenes. In addition, accessible multi-person sports video datasets lack fine- grained action annotations in both space and time, adding to the difficulty in fine- grained action analysis. To this end, we construct a new multi-person basketball sports video dataset named FineSports, which contains fine-grained semantic and spatial-temporal annotations on 10,000 NBA game videos, covering 52 fine-grained action types, 16,000 action instances, and 123,000 spatial-temporal bounding boxes. We also propose a new prompt-driven spatial-temporal action location approach called PoSTAL, composed of a prompt-driven target action encoder (PTA) and an action tube-specific detector (ATD) to directly generate target action tubes with fine-grained action types without any off-line proposal generation. Extensive experiments on the FineSports dataset demonstrate that PoSTAL outperforms state-of-the-art methods. Data and code are available at https://github.com/PKU-ICST-MIPL/FineSports_CVPR2024.Keywords
This publication has 33 references indexed in Scilit:
- An overview of Human Action Recognition in sports based on Computer VisionHeliyon, 2022
- Social Adaptive Module for Weakly-Supervised Group Activity Recognition2020
- Actions as Moving Points2020
- Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?2018
- AVA: A Video Dataset of Spatio-Temporally Localized Atomic Visual Actions2018
- Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset2017
- Multi-region Two-Stream R-CNN for Action Detection2016
- A Hierarchical Deep Temporal Model for Group Activity Recognition2016
- Finding action tubes2015
- ActivityNet: A large-scale video benchmark for human activity understanding2015