eduzhai > Applied Sciences > Engineering >

Generating Person Images with Appearance-aware Pose Stylizer

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 7 pages

Abstract: Generation of high-quality person images is challenging, due to thesophisticated entanglements among image factors, e.g., appearance, pose,foreground, background, local details, global structures, etc. In this paper,we present a novel end-to-end framework to generate realistic person imagesbased on given person poses and appearances. The core of our framework is anovel generator called Appearance-aware Pose Stylizer (APS) which generateshuman images by coupling the target pose with the conditioned person appearanceprogressively. The framework is highly flexible and controllable by effectivelydecoupling various complex person image factors in the encoding phase, followedby re-coupling them in the decoding phase. In addition, we present a newnormalization method named adaptive patch normalization, which enablesregion-specific normalization and shows a good performance when adopted inperson image generation model. Experiments on two benchmark datasets show thatour method is capable of generating visually appealing and realistic-lookingresults using arbitrary image and pose inputs.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×