MA2.L3: Contrastive Learning for Image & Video Data
Mon, 28 Oct, 10:30 - 12:00 Gulf Standard Time (UTC +4)
Location: Capital Suite - 16
Session Type: Lecture
Session Chair: Hak Gu Kim, Chung-Ang University
Track: Visual Artificial Intelligence
Click the to view the manuscript on IEEE Xplore Open Preview
Mon, 28 Oct, 10:30 - 10:48 Gulf Standard Time (UTC +4)
 

MA2.L3.1: Masked Momentum Contrastive Learning for Semantic Understanding by Observation
Jiantao Wu, University of Surrey, United Kingdom of Great Britain and Northern Ireland; Shentong Mo, Carnegie Mellon University, United States of America; Sara Atito, Zhenhua Feng, Josef Kittler, Syed Sameed Husain, Muhammad Awais, University of Surrey, United Kingdom of Great Britain and Northern Ireland
Mon, 28 Oct, 10:48 - 11:06 Gulf Standard Time (UTC +4)
 

MA2.L3.2: TAXES ARE ALL YOU NEED: INTEGRATION OF TAXONOMICAL HIERARCHY RELATIONSHIPS INTO THE CONTRASTIVE LOSS
Kiran Kokilepersaud, Yavuz Yarici, Mohit Prabhushankar, Ghassan AlRegib, Georgia Institute of Technology, United States of America
Mon, 28 Oct, 11:06 - 11:24 Gulf Standard Time (UTC +4)
 

MA2.L3.3: Imbalanced data robust online continual learning based on evolving class aware memory selection and built-in contrastive representation learning
rui yang, Emmanuel Dellandrea, école centrale de lyon, France; Matthieu Grard, siléane, France; Liming CHEN, école centrale de lyon, France
Mon, 28 Oct, 11:24 - 11:42 Gulf Standard Time (UTC +4)
 

MA2.L3.4: Conditional Past Experience Generation for Dark Continual Learning
Cheng Feng, Chaoliang Zhong, jie wang, jun sun, Fujitsu R&D Center, Co., LTD, China; Yasuto Yokota, Fujitsu LTD, Japan
Mon, 28 Oct, 11:42 - 12:00 Gulf Standard Time (UTC +4)
 

MA2.L3.5: UNSUPERVISED DOMAIN ADAPTIVE SEMANTIC SEGMENTATION BASED ON CLIP-GUIDED PROTOTYPICAL CONTRASTIVE LEARNING
Kebin Liu, Chuang Zhu, Beijing University of Posts and Telecommunications, China