FDA-GAN: Flow-based Dual Attention GAN for Human Pose Transfer

TL;DR

FDA-GAN introduces a dual attention mechanism with deformable local and flow similarity attention for improved human pose transfer, effectively preserving appearance details and ensuring pose consistency, outperforming existing models.

Contribution

The paper proposes FDA-GAN, a novel flow-based dual attention GAN that enhances feature fusion for human pose transfer, addressing appearance detail preservation and pose consistency.

Findings

Outperforms state-of-the-art models on iPER and DeepFashion datasets.

Effectively preserves appearance details in pose transfer.

Maintains pose and global position consistency.

Abstract

Human pose transfer aims at transferring the appearance of the source person to the target pose. Existing methods utilizing flow-based warping for non-rigid human image generation have achieved great success. However, they fail to preserve the appearance details in synthesized images since the spatial correlation between the source and target is not fully exploited. To this end, we propose the Flow-based Dual Attention GAN (FDA-GAN) to apply occlusion- and deformation-aware feature fusion for higher generation quality. Specifically, deformable local attention and flow similarity attention, constituting the dual attention mechanism, can derive the output features responsible for deformable- and occlusion-aware fusion, respectively. Besides, to maintain the pose and global position consistency in transferring, we design a pose normalization network for learning adaptive normalization from the target pose to the source person. Both qualitative and quantitative results show that our method outperforms state-of-the-art models in public iPER and DeepFashion datasets.