STGAN: A Unified Selective Transfer Network for Arbitrary Image Attribute Editing
This work addresses image editing challenges for computer vision applications, offering an incremental improvement over existing encoder-decoder and GAN-based approaches.
The paper tackles the problem of arbitrary image attribute editing, where existing methods suffer from blurry results or weakened manipulation ability, by proposing STGAN, a selective transfer network that improves both attribute manipulation accuracy and perceptual quality, achieving favorable performance against state-of-the-art methods in tasks like facial attribute editing and season translation.
Arbitrary attribute editing generally can be tackled by incorporating encoder-decoder and generative adversarial networks. However, the bottleneck layer in encoder-decoder usually gives rise to blurry and low quality editing result. And adding skip connections improves image quality at the cost of weakened attribute manipulation ability. Moreover, existing methods exploit target attribute vector to guide the flexible translation to desired target domain. In this work, we suggest to address these issues from selective transfer perspective. Considering that specific editing task is certainly only related to the changed attributes instead of all target attributes, our model selectively takes the difference between target and source attribute vectors as input. Furthermore, selective transfer units are incorporated with encoder-decoder to adaptively select and modify encoder feature for enhanced attribute editing. Experiments show that our method (i.e., STGAN) simultaneously improves attribute manipulation accuracy as well as perception quality, and performs favorably against state-of-the-arts in arbitrary facial attribute editing and season translation.