Progressive Pose Attention Transfer for Person Image Generation

This paper proposes a new generative adversarial network for pose transfer, i.e., transferring the pose of a given person to a target pose. The generator of the network comprises a sequence of Pose-Attentional Transfer Blocks that each transfers certain regions it attends to, generating the person image progressively. Compared with those in previous works, our generated person images possess better appearance consistency and shape consistency with the input images, thus significantly more realistic-looking. The efficacy and efficiency of the proposed network are validated both qualitatively and quantitatively on Market-1501 and DeepFashion. Furthermore, the proposed architecture can generate training images for person re-identification, alleviating data insufficiency. Codes and models are available at:

PDF Abstract CVPR 2019 PDF CVPR 2019 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Pose Transfer Deep-Fashion Progressive Pose Attention SSIM 0.773 # 5
IS 3.209 # 7
DS 0.976 # 1
PCKh 0.96 # 2
Retrieval Top10 Recall 17.84 # 3
Pose Transfer Market-1501 Progressive Pose Attention DS 0.74 # 1
IS 3.323 # 3
PCKh 0.94 # 1
SSIM 0.311 # 3
mask-IS 3.773 # 2
mask-SSIM 0.811 # 3


No methods listed for this paper. Add relevant methods here