WAS-VTON: Warping Architecture Search for Virtual Try-on Network

TL;DR

WAS-VTON introduces a neural architecture search approach to find clothing category-specific warping networks for virtual try-on, resulting in more natural and accurate try-on images.

Contribution

The paper proposes a novel NAS-based framework that automatically designs warping and fusion modules tailored for different clothing categories in virtual try-on.

Findings

Outperforms previous fixed-architecture methods in naturalness and accuracy

Achieves better clothing-person alignment through optimized warping networks

Produces more seamless and realistic virtual try-on results

Abstract

Despite recent progress on image-based virtual try-on, current methods are constraint by shared warping networks and thus fail to synthesize natural try-on results when faced with clothing categories that require different warping operations. In this paper, we address this problem by finding clothing category-specific warping networks for the virtual try-on task via Neural Architecture Search (NAS). We introduce a NAS-Warping Module and elaborately design a bilevel hierarchical search space to identify the optimal network-level and operation-level flow estimation architecture. Given the network-level search space, containing different numbers of warping blocks, and the operation-level search space with different convolution operations, we jointly learn a combination of repeatable warping cells and convolution operations specifically for the clothing-person alignment. Moreover, a NAS-Fusion Module is proposed to synthesize more natural final try-on results, which is realized by leveraging particular skip connections to produce better-fused features that are required for seamlessly fusing the warped clothing and the unchanged person part. We adopt an efficient and stable one-shot searching strategy to search the above two modules. Extensive experiments demonstrate that our WAS-VTON significantly outperforms the previous fixed-architecture try-on methods with more natural warping results and virtual try-on results.