IVMSP-L9.4
DAMP: DISTRIBUTION-AWARE MAGNITUDE PRUNING FOR BUDGET-SENSITIVE GRAPH CONVOLUTIONAL NETWORKS
Hichem Sahbi, CNRS Sorbonne University, France
Session:
IVMSP-L9: Deep learning theory Lecture
Track:
Image, Video, and Multidimensional Signal Processing
Location:
Room 105
Presentation Time:
Thu, 18 Apr, 09:20 - 09:40 (UTC +9)
Session Co-Chairs:
Hichem Sahbi, CNRS Sorbonne University and Nicola Conci , University of Trento
Session IVMSP-L9
IVMSP-L9.1: Dual Rank-1 Tensor Attention Module for Convolutional Neural Networks
Baihong Lin, UESTC, China; Hanxing Chi, Zengrong Lin, Jun Hu, Liang Wang, Sun Yat-sen University, China; Jianxiao Zou, Shicai Fan, UESTC, China
IVMSP-L9.2: DEEP NEIGHBOR LAYER AGGREGATION FOR LIGHTWEIGHT SELF-SUPERVISED MONOCULAR DEPTH ESTIMATION
Boya Wang, Harbin Institute of Technology, China; Shuo Wang, Shanghai Institute of Microsystem and Information Technology, China; Dong Ye, Ziwen Dou, Harbin Institute of Technology, China
IVMSP-L9.3: SELF-SUPERVISED MULTI-SCALE HIERARCHICAL REFINEMENT METHOD FOR JOINT LEARNING OF OPTICAL FLOW AND DEPTH
Rokia Abdein, Xuezhi Xiang, Yiming Chen, Harbin Engineering University, China; Mingliang Zhai, Nanjing University of Posts and Telecommunications, China; Abdulmotaleb El Saddik, University of Ottawa, Canada
IVMSP-L9.4: DAMP: DISTRIBUTION-AWARE MAGNITUDE PRUNING FOR BUDGET-SENSITIVE GRAPH CONVOLUTIONAL NETWORKS
Hichem Sahbi, CNRS Sorbonne University, France
IVMSP-L9.5: Lighting Image/Video Style Transfer Methods by Iterative Channel Pruning
Kexin Wu, Jilin University, China; Fan Tang, Institute of Computing Technology (ICT), CAS China, China; Ning Liu, Midea Group, China; Oliver Deussen, University of Konstanz, Germany; Thi-ngoc-hanh Le, National Cheng Kung University, Taiwan; Weiming Dong, Institute of Automation, Chinese Academy of Sciences, China; Tong-yee Lee, National Cheng Kung University, Taiwan
IVMSP-L9.6: The Devil is in Details: Delving into Lite FFN Design for Vision Transformers
Zhiyang Chen, Yousong Zhu, Zhaowen Li, Fan Yang, Chaoyang Zhao, Jinqiao Wang, Ming Tang, Institute of Automation, Chinese Academy of Sciences, China
Contacts