Improving Detail in Pluralistic Image Inpainting with Feature Dequantization
This addresses a critical challenge in image inpainting for applications like editing and object removal, though it is incremental as it builds on existing VQGAN methods.
The paper tackles the degradation of detail quality in pluralistic image inpainting caused by feature quantization in VQGAN-based models, proposing a Feature Dequantization Module (FDM) that significantly enhances detail quality with negligible overhead.
Pluralistic Image Inpainting (PII) offers multiple plausible solutions for restoring missing parts of images and has been successfully applied to various applications including image editing and object removal. Recently, VQGAN-based methods have been proposed and have shown that they significantly improve the structural integrity in the generated images. Nevertheless, the state-of-the-art VQGAN-based model PUT faces a critical challenge: degradation of detail quality in output images due to feature quantization. Feature quantization restricts the latent space and causes information loss, which negatively affects the detail quality essential for image inpainting. To tackle the problem, we propose the FDM (Feature Dequantization Module) specifically designed to restore the detail quality of images by compensating for the information loss. Furthermore, we develop an efficient training method for FDM which drastically reduces training costs. We empirically demonstrate that our method significantly enhances the detail quality of the generated images with negligible training and inference overheads.