Attention-guided GANs for human pose transfer

Jinsong Zhang; Yuyang Zhao; Kun Li; Yebin Liu; Jingyu Yang; Qionghai Dai

doi:10.1117/12.2538638

Translator Disclaimer

18 November 2019 Attention-guided GANs for human pose transfer

Jinsong Zhang, Yuyang Zhao, Kun Li, Yebin Liu, Jingyu Yang, Qionghai Dai

Author Affiliations +

Proceedings Volume 11187, Optoelectronic Imaging and Multimedia Technology VI; 111870W (2019) https://doi.org/10.1117/12.2538638
Event: SPIE/COS Photonics Asia, 2019, Hangzhou, China

Abstract

This paper presents a novel generative adversarial network for the task of human pose transfer, which aims at transferring the pose of a given person to a target pose. In order to deal with pixel-to-pixel misalignment due to the pose differences, we introduce an attention mechanism and propose Pose-Guided Attention Blocks. With these blocks, the generator can learn how to transfer the details from the conditional image to the target image based on the target pose. Our network can make the target pose truly guide the transfer of features. The effectiveness of the proposed network is validated on DeepFasion and Market-1501 datasets. Compared with state-of-the-art methods, our generated images are more realistic with better facial details.

Citation Download Citation

Jinsong Zhang, Yuyang Zhao, Kun Li, Yebin Liu, Jingyu Yang, and Qionghai Dai "Attention-guided GANs for human pose transfer", Proc. SPIE 11187, Optoelectronic Imaging and Multimedia Technology VI, 111870W (18 November 2019); https://doi.org/10.1117/12.2538638

ACCESS THE FULL ARTICLE