Zhao, ZiyiKiciroglu, SenaVinzant, HuguesCheng, YuanKatircioglu, IsinsuSalzmann, MathieuFua, Pascal2022-11-242022-11-242022-11-2410.1007/978-3-031-26316-3_12https://infoscience.epfl.ch/handle/20.500.14299/192737WOS:001000822000012Unsupervised self-rehabilitation exercises and physical training can cause serious injuries if performed incorrectly. We introduce a learning-based framework that identifies the mistakes made by a user and proposes corrective measures for easier and safer individual training. Our framework does not rely on hard-coded, heuristic rules. Instead, it learns them from data, which facilitates its adaptation to specific user needs. To this end, we use a Graph Convolutional Network (GCN) architecture acting on the user's pose sequence to model the relationship between the the body joints trajectories. To evaluate our approach, we introduce a dataset with 3 different physical exercises. Our approach yields 90.9% mistake identification accuracy and successfully corrects 94.2% of the mistakes.physical exercise supervisionhuman poseaction recognitiontext::conference output::conference proceedings::conference paper