MQ-L.C1.6

I SAW: A SELF-ATTENTION WEIGHTED METHOD FOR EXPLANATION OF VISUAL TRANSFORMERS

Rupayan Mallick, Jenny Benois-Pineau, Akka Zemmari, University of Bordeaux, France
Session:
Image / Video Interpretation and Understanding
Lecture
Track:
Image and Video Analysis, Synthesis, and Retrieval
Location:
Room C1
Presentation Time:
Mon, 17 Oct, 21:35 - 21:50 China Standard Time (UTC +8)
Mon, 17 Oct, 15:35 - 15:50 Central European Time (UTC +2)
Mon, 17 Oct, 13:35 - 13:50 UTC
Mon, 17 Oct, 09:35 - 09:50 Eastern Time (UTC -4)
Session Chair:
Aydin Alatan, METU
Presentation
Discussion
Resources
No resources available.
Session MQ-L.C1
MQ-L.C1.1: HYPERBOLIC SPATIAL TEMPORAL GRAPH CONVOLUTIONAL NETWORKS
Abdelrahman Mostafa, Wei Peng, Guoying Zhao, University of Oulu, Finland
MQ-L.C1.2: GANZZLE: REFRAMING JIGSAW PUZZLE SOLVING AS A RETRIEVAL TASK USING A GENERATIVE MENTAL IMAGE
Davide Talon, Alessio Del Bue, Stuart James, Istituto Italiano di Tecnologia, Italy
MQ-L.C1.3: FLOW-PATH FITTING FROM IMAGES WITH FOURIER BASIS FOR RIVER HEALTH ASSESSMENT
Yuki Takahashi, Shogo Muramatsu, Hiroyasu Yasuda, Kiyoshi Hayasaka, Niigata University, Japan; Yu Otake, Tohoku University, Japan
MQ-L.C1.4: Panoptic-DeepLab-DVA: Improving Panoptic DeepLab with Dual Value Attention and instance Boundary Aware Regression
Qingfeng Liu, Mostafa El-Khamy, Samsung USA, United States of America
MQ-L.C1.5: MVMO: A MULTI-OBJECT DATASET FOR WIDE BASELINE MULTI-VIEW SEMANTIC SEGMENTATION
Aitor Alvarez-Gila, TECNALIA - Basque Research and Technology Alliance (BRTA) / Computer Vision Center, Spain; Joost van de Weijer, Computer Vision Center, Spain; Yaxing Wang, Nankai University, China; Estibaliz Garrote, TECNALIA - Basque Research and Technology Alliance (BRTA), Spain
MQ-L.C1.6: I SAW: A SELF-ATTENTION WEIGHTED METHOD FOR EXPLANATION OF VISUAL TRANSFORMERS
Rupayan Mallick, Jenny Benois-Pineau, Akka Zemmari, University of Bordeaux, France