MA2.L4.4
CASeg: CLIP-based Action Segmentation with learnable text prompt
Suyuan Huang, Beihang University, China; Haoxin Zhang, Shuxing Technology (Beijing) Co., Ltd., China; Yanyu Xu, The Chinese University of Hong Kong, China; Yan Gao, Yao Hu, Shuxing Technology (Beijing) Co., Ltd., China; Zengchang Qin, Beihang University, China
Session:
MA2.L4: Image & Video Interpretation & Understanding - I Lecture
Track:
Image and Video Analysis, Synthesis, and Retrieval
Location:
Lecture Room 4
Presentation Time:
Mon, 28 Oct, 11:24 - 11:42 Gulf Standard Time (UTC +4)
Session Chair:
Zengchang Qin, Beihang University
Session MA2.L4
MA2.L4.1: CROSS-ACTION CROSS-SUBJECT SKELETON ACTION RECOGNITION VIA SIMULTANEOUS ACTION-SUBJECT LEARNING WITH TWO-STEP FEATURE REMOVAL
Yu Mitsuzumi, Akisato Kimura, NTT Corporation, Japan; Go Irie, Tokyo University of Science / NTT Corporation, Japan; Atsushi Nakazawa, Okayama University / Kyoto University, Japan
MA2.L4.2: RETHINKING TEMPORAL SELF-SIMILARITY FOR REPETITIVE ACTION COUNTING
Yanan Luo, Jinhui Yi, University of Bonn, Germany; Yazan Abu Farha, Birzeit University, Palestine, State of; Moritz Wolter, University of Bonn, Germany; Juergen Gall, University of Bonn, Lamarr Institute for Machine Learning and Artificial Intelligence, Germany
MA2.L4.3: Subgroups for Detection Transformer
Tharsan Senthivel, PMU/ETIS, France; Ngoc-Son Vu, ETIS, France
MA2.L4.4: CASeg: CLIP-based Action Segmentation with learnable text prompt
Suyuan Huang, Beihang University, China; Haoxin Zhang, Shuxing Technology (Beijing) Co., Ltd., China; Yanyu Xu, The Chinese University of Hong Kong, China; Yan Gao, Yao Hu, Shuxing Technology (Beijing) Co., Ltd., China; Zengchang Qin, Beihang University, China
MA2.L4.5: Prompt Performance Prediction for Image Generation
Nicolas BIZZOZZERO, done as independent, France; Ihab BENDIDI, Ecole Normale Supérieure, France; Olivier RISSER-MAROIX, Artinity, France
Contacts