Enhancing Robustness to Domain Shifts in Vision Transformer Models

Home

Discover

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and Machine Learning and more.Subscribe

CC BY

Source arXiv

Computer Vision

Enhancing Robustness to Domain Shifts in Vision Transformer Models

Optimizing Relevance Maps of Vision Transformers Improves Robustness

Visual classification models often rely mostly on the image background, neglecting the foreground, which hurts their robustness to distribution changes.

To alleviate this shortcoming, we propose to monitor the model s relevancy signal and manipulate it such that the model is focused on the foreground object.

This is done as a finetuning step, involving relatively few samples consisting of pairs of images and their associated foreground masks.

When applied to vision transformer (vit)models, a marked improvement in robustness to domain shifts is observed.

Authors

Hila Chefer, Idan Schwartz, Lior Wolf

◐ Recommended Follows

◐ Latest Activity