1 paper
Liyang Peng, Sihan Zhu, Yunjie Guo
Action recognition and localization in complex, untrimmed videos remain a formidable challenge in computer vision, largely due to the limitations of existing methods in capturing f…