Visual Prompting via Image Inpainting

Amir Bar, Yossi Gandelsman, Trevor Darrell, Amir Globerson, Alexei A Efros

Keywords: Computer Vision, Self-Supervised Learning, Visual Prompting

TL;DR: We study Visual Prompting and explore how to adapt models to variety of downstream tasks without finetuning or weight modification.

Abstract: How does one adapt a pre-trained visual model to novel downstream tasks without task-specific finetuning or any model modification? Inspired by prompting in NLP, this paper investigates visual prompting: given input-output image example(s) of a new task at test time and a new input image, the goal is to automatically produce the output image, consistent with the given examples. We show that posing this problem as simple image inpainting -- literally just filling in a hole in a concatenated visual prompt image -- turns out to be surprisingly effective, provided that the inpainting algorithm has been trained on the right data. We train masked auto-encoders on a new dataset that we curated -- 88k unlabeled figures from academic papers sources on Arxiv. We apply visual prompting to these pretrained models and demonstrate results on various downstream image-to-image tasks, including foreground segmentation, single object detection, colorization, edge detection, etc. Project page: https://yossigandelsman.github.io/visual_prompt

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:2209.00647/code)