CuDi3D: Curvilinear displacement based approach for online 3D action detection

• Published in: Computer vision and image understanding Vol. 174; pp. 57 - 69
• Main Authors: Boulahia, Said Yacine, Anquetil, Eric, Multon, Franck, Kulpa, Richard
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.09.2018; Elsevier
• Subjects: Computer Science; Computer Vision and Pattern Recognition; Curvilinear displacement; Graphics; Human action detection; Human-Computer Interaction; Machine Learning; Multimedia; Online action recognition; Online segmentation; Skeleton data stream; Skeleton-based approach; Online action recognition; Human action detection; Online segmentation; Skeleton-based approach; Skeleton data stream; Curvilinear displacement

•A step by step approach that addresses the OAD problem in a transparent manner.•A novel and robust online segmentation paradigm for skeleton-based action detection.•Addressing temporal variability with curvilinear windows.•Addressing inter-class spatial variability using multiple classifiers.•Addre...