Pose-Guided High-Resolution Appearance Transfer via Progressive Training
This addresses the challenge of realistic human image generation for applications like garment transfer and video synthesis, representing a novel method for a known bottleneck.
The paper tackles the problem of transferring a reference person's appearance to a target pose at high resolution (1024x1024) without using 3D models, achieving high-quality results that preserve garment textures and handle dis-occlusions.
We propose a novel pose-guided appearance transfer network for transferring a given reference appearance to a target pose in unprecedented image resolution (1024 * 1024), given respectively an image of the reference and target person. No 3D model is used. Instead, our network utilizes dense local descriptors including local perceptual loss and local discriminators to refine details, which is trained progressively in a coarse-to-fine manner to produce the high-resolution output to faithfully preserve complex appearance of garment textures and geometry, while hallucinating seamlessly the transferred appearances including those with dis-occlusion. Our progressive encoder-decoder architecture can learn the reference appearance inherent in the input image at multiple scales. Extensive experimental results on the Human3.6M dataset, the DeepFashion dataset, and our dataset collected from YouTube show that our model produces high-quality images, which can be further utilized in useful applications such as garment transfer between people and pose-guided human video generation.