TA2.L3: Augmentation for Image & Video Learning
Tue, 29 Oct, 10:30 - 12:00 Gulf Standard Time (UTC +4)
Location: Capital Suite - 15
Session Type: Lecture
Session Chair: Xiaohong Liu, Shanghai Jiao Tong University
Track: Visual Artificial Intelligence
Click the to view the manuscript on IEEE Xplore Open Preview
Tue, 29 Oct, 10:30 - 10:48 Gulf Standard Time (UTC +4)
 

TA2.L3.1: ECAP: EXTENSIVE CUT-AND-PASTE AUGMENTATION FOR UNSUPERVISED DOMAIN ADAPTIVE SEMANTIC SEGMENTATION
Erik Brorsson, Volvo Group, Chalmers University of Technology, Sweden; Knut Åkesson, Lennart Svensson, Chalmers University of Technology, Sweden; Kristofer Bengtsson, Volvo Group, Sweden
Tue, 29 Oct, 10:48 - 11:06 Gulf Standard Time (UTC +4)
 

TA2.L3.2: INTELLIGENT MULTI-VIEW TEST TIME AUGMENTATION
Efe Ozturk, Mohit Prabhushankar, Ghassan AlRegib, Georgia Institute of Technology, United States of America
Tue, 29 Oct, 11:06 - 11:24 Gulf Standard Time (UTC +4)
 

TA2.L3.3: SUPERPIXEL MIXING: A DATA AUGMENTATION TECHNIQUE FOR ROBUST DEEP VISUAL RECOGNITION MODELS
Danyang Sun, Fadi Dornaika, University of the Basque Country, Spain; Vinh Hoang, Ho Chi Minh City Open University, Viet Nam; Nagore Barrena, University of the Basque Country, Spain
Tue, 29 Oct, 11:24 - 11:42 Gulf Standard Time (UTC +4)
 Best Paper Candidate

TA2.L3.4: DIVERSIFIED TASK AUGMENTATION WITH REDUNDANCY REDUCTION FOR CROSS-DOMAIN FEW-SHOT LEARNING
Ling Yue, Lin Feng, Qiuping Shuai, Lingxiao Xu, Zihao Li, Sichuan Normal University, China
Tue, 29 Oct, 11:42 - 12:00 Gulf Standard Time (UTC +4)
 

TA2.L3.5: SEMI-SUPERVISED 3D OBJECT DETECTION WITH CHANNEL AUGMENTATION USING TRANSFORMATION EQUIVARIANCE
Minju Kang, KAIST, LG Electronics, Korea, Republic of; Taehun Kong, KAIST, Korea, Republic of; Tae-Kyun Kim, KAIST, Imperial College London, Korea, Republic of