Dressing in the Wild by Watching Dance Videos
Xin Dong, Fuwei Zhao, Zhenyu Xie, Xijin Zhang, Daniel K. Du, Min, Zheng, Xiang Long, Xiaodan Liang, Jianchao Yang

TL;DR
This paper introduces wFlow, a novel generative network that improves virtual try-on in real-world scenes by combining pixel and vertex flows, trained on a large-scale dance video dataset, achieving realistic garment transfer without paired data.
Contribution
The paper presents a new network architecture, wFlow, that effectively combines pixel and vertex flows for in-the-wild garment transfer, and introduces Dance50k, a large-scale dataset for self-supervised training.
Findings
wFlow outperforms existing methods in realism and accuracy.
Combining pixel and vertex flows enhances handling of loose garments and challenging poses.
Dance50k enables training without paired images, reducing data collection effort.
Abstract
While significant progress has been made in garment transfer, one of the most applicable directions of human-centric image generation, existing works overlook the in-the-wild imagery, presenting severe garment-person misalignment as well as noticeable degradation in fine texture details. This paper, therefore, attends to virtual try-on in real-world scenes and brings essential improvements in authenticity and naturalness especially for loose garment (e.g., skirts, formal dresses), challenging poses (e.g., cross arms, bent legs), and cluttered backgrounds. Specifically, we find that the pixel flow excels at handling loose garments whereas the vertex flow is preferred for hard poses, and by combining their advantages we propose a novel generative network called wFlow that can effectively push up garment transfer to in-the-wild context. Moreover, former approaches require paired images for…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
Topics3D Shape Modeling and Analysis · Generative Adversarial Networks and Image Synthesis · Image Enhancement Techniques
