Transferring Knowledge with Attention Distillation for Multi-Domain Image-to-Image Translation

08/17/2021
by   Runze Li, et al.
6

Gradient-based attention modeling has been used widely as a way to visualize and understand convolutional neural networks. However, exploiting these visual explanations during the training of generative adversarial networks (GANs) is an unexplored area in computer vision research. Indeed, we argue that this kind of information can be used to influence GANs training in a positive way. For this reason, in this paper, it is shown how gradient based attentions can be used as knowledge to be conveyed in a teacher-student paradigm for multi-domain image-to-image translation tasks in order to improve the results of the student architecture. Further, it is demonstrated how "pseudo"-attentions can also be employed during training when teacher and student networks are trained on different domains which share some similarities. The approach is validated on multi-domain facial attributes transfer and human expression synthesis showing both qualitative and quantitative results.

READ FULL TEXT

page 2

page 4

page 6

page 7

page 8

research
04/30/2021

Semantic Relation Preserving Knowledge Distillation for Image-to-Image Translation

Generative adversarial networks (GANs) have shown significant potential ...
research
05/25/2022

Region-aware Knowledge Distillation for Efficient Image-to-Image Translation

Recent progress in image-to-image translation has witnessed the success ...
research
03/07/2020

Distilling portable Generative Adversarial Networks for Image Translation

Despite Generative Adversarial Networks (GANs) have been widely used in ...
research
05/28/2019

Video-to-Video Translation for Visual Speech Synthesis

Despite remarkable success in image-to-image translation that celebrates...
research
02/17/2023

Transformer-based Generative Adversarial Networks in Computer Vision: A Comprehensive Survey

Generative Adversarial Networks (GANs) have been very successful for syn...
research
04/22/2023

BiTrackGAN: Cascaded CycleGANs to Constraint Face Aging

With the increased accuracy of modern computer vision technology, many a...
research
01/18/2019

Learning Spatial Pyramid Attentive Pooling in Image Synthesis and Image-to-Image Translation

Image synthesis and image-to-image translation are two important generat...

Please sign up or login with your details

Forgot password? Click here to reset