LIM, JUNGJUN KANG, DONG HOON RYU, KANGHYUN HONG, JE HYEONG 2025-11-04T08:30:19Z 2025-11-04T08:30:19Z 2025-10-31 2025-10 https://pubs.kist.re.kr/handle/201004/153401 Most existing approaches to predicting the number of repetitive actions in videos focus on improving model accuracy, but overlook important issues such as robustness to changes in human body size and and occlusion of human body parts in videos. To achieve robustness to changes in human size and and occlusion of human body in videos, we propose a novel network, Skeleton-based dual-stream Spatio-temporal Transformer for Repetitive Action Counting (SSTRAC) using videos, which reconstructs defective human skeletons as a preprocessing step, and then encodes the spatial and temporal information of repetitive actions into the per-frame embeddings through the dual-stream spatio-temporal transformer. To capture both high and low frequency actions in short and long videos, the per-frame embeddings are abstracted in the form of a multi-scale self-attention matrix. In the final step, the period predictor estimates a density map, which provides the number of repetitive actions in each video. We performed extensive experiments by comparing the proposed model with other recent state-of-the art models. The experimental results demonstrate the superiority of our model in terms of robustness to changes in human size and occlusion of human body parts in videos. Codes and models are available at https://github.com/imjjun/SSTRAC_public English Institute of Electrical and Electronics Engineers Inc. SSTRAC: Skeleton-Based Dual-Stream Spatio-Temporal Transformer for Repetitive Action Counting in Videos Article 10.1109/ACCESS.2025.3624029 1 IEEE Access, v.13, pp.184046 - 184058 IEEE Access 13 184046 184058 Y scie scopus 001606717700026 Computer Science, Information Systems Engineering, Electrical & Electronic Telecommunications Computer Science Engineering Telecommunications Article Videos Skeleton Transformers Robustness Data models Predictive models Cameras Art Solid modeling Silicon Repetitive action counting skeleton human size occlusion spatio-temporal multi-scale density map