WaveFace: Authentic Face Restoration with Efficient Frequency Recovery
This work addresses efficiency and identity preservation issues in face restoration for applications like image enhancement, but it is incremental as it builds on existing diffusion model approaches.
The paper tackled the problems of slow speed and poor identity preservation in diffusion models for blind face restoration by proposing WaveFace, which uses wavelet transformation to separately handle low- and high-frequency components, resulting in 10x faster efficiency and improved authenticity on benchmark datasets.
Although diffusion models are rising as a powerful solution for blind face restoration, they are criticized for two problems: 1) slow training and inference speed, and 2) failure in preserving identity and recovering fine-grained facial details. In this work, we propose WaveFace to solve the problems in the frequency domain, where low- and high-frequency components decomposed by wavelet transformation are considered individually to maximize authenticity as well as efficiency. The diffusion model is applied to recover the low-frequency component only, which presents general information of the original image but 1/16 in size. To preserve the original identity, the generation is conditioned on the low-frequency component of low-quality images at each denoising step. Meanwhile, high-frequency components at multiple decomposition levels are handled by a unified network, which recovers complex facial details in a single step. Evaluations on four benchmark datasets show that: 1) WaveFace outperforms state-of-the-art methods in authenticity, especially in terms of identity preservation, and 2) authentic images are restored with the efficiency 10x faster than existing diffusion model-based BFR methods.