Human Pose Transfer by Adaptive Hierarchical Deformation

TL;DR

This paper introduces an adaptive hierarchical deformation network for human pose transfer that effectively preserves style and shape details, outperforming existing methods in quality and efficiency.

Contribution

The proposed model employs a two-level hierarchical deformation with semantic guidance and gated convolutions, enabling better detail preservation and faster convergence.

Findings

Achieves more consistent hair, face, and clothes compared to state-of-the-art methods.

Uses fewer parameters and converges faster.

Can be applied to clothing texture transfer.

Abstract

Human pose transfer, as a misaligned image generation task, is very challenging. Existing methods cannot effectively utilize the input information, which often fail to preserve the style and shape of hair and clothes. In this paper, we propose an adaptive human pose transfer network with two hierarchical deformation levels. The first level generates human semantic parsing aligned with the target pose, and the second level generates the final textured person image in the target pose with the semantic guidance. To avoid the drawback of vanilla convolution that treats all the pixels as valid information, we use gated convolution in both two levels to dynamically select the important features and adaptively deform the image layer by layer. Our model has very few parameters and is fast to converge. Experimental results demonstrate that our model achieves better performance with more consistent hair, face and clothes with fewer parameters than state-of-the-art methods. Furthermore, our method can be applied to clothing texture transfer.