Test-Time Adaptation for Out-of-distributed Image Inpainting

TL;DR

AdaFill is a test-time adaptation method for image inpainting that leverages internal image priors to improve results on out-of-distribution images, outperforming pre-trained models.

Contribution

It introduces a simple test-time adaptation approach for image inpainting that exploits internal image priors, enhancing performance on out-of-distribution images.

Findings

AdaFill outperforms other models on various out-of-distribution images.

ZeroFill, a non-pre-trained model, sometimes surpasses pre-trained models.

Test-time adaptation improves inpainting quality on unseen image distributions.

Abstract

Deep learning-based image inpainting algorithms have shown great performance via powerful learned prior from the numerous external natural images. However, they show unpleasant results on the test image whose distribution is far from the that of training images because their models are biased toward the training images. In this paper, we propose a simple image inpainting algorithm with test-time adaptation named AdaFill. Given a single out-of-distributed test image, our goal is to complete hole region more naturally than the pre-trained inpainting models. To achieve this goal, we treat remained valid regions of the test image as another training cues because natural images have strong internal similarities. From this test-time adaptation, our network can exploit externally learned image priors from the pre-trained features as well as the internal prior of the test image explicitly. Experimental results show that AdaFill outperforms other models on the various out-of-distribution test images. Furthermore, the model named ZeroFill, that are not pre-trained also sometimes outperforms the pre-trained models.