Indexed by:期刊论文
Date of Publication:2020-11-06
Journal:NEUROCOMPUTING
Volume:413
Page Number:360-367
ISSN No.:0925-2312
Key Words:Action recognition; Figure Skating Dataset; Fine-grained sports content analysis; Keyframe based temporal segment network
Abstract:Action recognition is an important and challenging problem in video analysis. Although the past decade has witnessed progress in action recognition with the development of deep learning, such process has been slow in competitive sports content analysis. To promote the research on action recognition from competitive sports video clips, we introduce a Figure Skating Dataset (FSD-10) for fine-grained sports content analysis. To this end, we collect 1484 clips from the worldwide figure skating championships in 2017-2018, which consist of 10 different actions in men/ladies programs. Each clip is at a rate of 30 frames per second with resolution 1080 x 720, which are annotated by experts. To build a baseline for action recognition in figure skating, we evaluate state-of-the-art action recognition methods on FSD-10. Motivated by the idea that domain knowledge is of great concern in sports field, we propose a key-frame based temporal segment network (KTSN) for classification and achieve remarkable performance. Experimental results demonstrate that FSD-10 is an ideal dataset for benchmarking action recognition algorithms, as it requires to accurately extract action motions rather than action poses. We hope FSD-10, which is designed to have a large collection of finegrained actions, can serve as a new challenge to develop more robust and advanced action recognition models. (C) 2020 Elsevier B.V. All rights reserved.
Associate Professor
Supervisor of Master's Candidates
Gender:Male
Alma Mater:大连理工大学
Degree:Doctoral Degree
School/Department:创新创业学院
Discipline:Computer Applied Technology
Open time:..
The Last Update Time:..