LC-VTON: Length Controllable Virtual Try-On Network

Image-based virtual try-on provides customers with convenient online clothes selections by transferring garments onto a reference person. Despite the emergence of several solutions to generate photo-realistic images and adapt complex poses, controlling clothing length remains challenge. We argue that reconstruction did not consider information, which results in being uncontrollable most methods. To overcome this limitation, novel clothing-agnostic person representation is proposed, eliminates information quantifies as numerical value. A new segmentation generator designed predict maps any conditioned on representation. Moreover, we correct two inaccurate labels, enables our model wider range garment interactions images, such top tucked into or worn over bottom, well bottom separately without intersecting. Extensive experiments demonstrate method achieves goal continuous control generates fine details outperform baseline methods terms quantitative qualitative metrics.