Beyond Imperfections: A Conditional Inpainting Approach for End-to-End Artifact Removal in VTON and Pose Transfer

TL;DR

This paper presents a new end-to-end inpainting method tailored for artifact removal in virtual try-on and pose transfer images, significantly improving visual quality and setting a new standard in the field.

Contribution

It introduces the first end-to-end framework for artifact removal in VTON and pose transfer, along with a specialized dataset for training and evaluation.

Findings

Effective artifact removal demonstrated in experiments

Significant enhancement in image aesthetics

Sets new benchmark in visual quality

Abstract

Artifacts often degrade the visual quality of virtual try-on (VTON) and pose transfer applications, impacting user experience. This study introduces a novel conditional inpainting technique designed to detect and remove such distortions, improving image aesthetics. Our work is the first to present an end-to-end framework addressing this specific issue, and we developed a specialized dataset of artifacts in VTON and pose transfer tasks, complete with masks highlighting the affected areas. Experimental results show that our method not only effectively removes artifacts but also significantly enhances the visual quality of the final images, setting a new benchmark in computer vision and image processing.