RestoreFormer: High-Quality Blind Face Restoration from Undegraded Key-Value Pairs
This work addresses the problem of restoring high-quality face images from degraded inputs for applications in computer vision, representing an incremental improvement over existing methods.
The authors tackled blind face restoration from unknown degradations by proposing RestoreFormer, which uses fully-spatial attentions and high-quality key-value pairs, achieving superior results on synthetic and real-world datasets with better visual quality.
Blind face restoration is to recover a high-quality face image from unknown degradations. As face image contains abundant contextual information, we propose a method, RestoreFormer, which explores fully-spatial attentions to model contextual information and surpasses existing works that use local operators. RestoreFormer has several benefits compared to prior arts. First, unlike the conventional multi-head self-attention in previous Vision Transformers (ViTs), RestoreFormer incorporates a multi-head cross-attention layer to learn fully-spatial interactions between corrupted queries and high-quality key-value pairs. Second, the key-value pairs in ResotreFormer are sampled from a reconstruction-oriented high-quality dictionary, whose elements are rich in high-quality facial features specifically aimed for face reconstruction, leading to superior restoration results. Third, RestoreFormer outperforms advanced state-of-the-art methods on one synthetic dataset and three real-world datasets, as well as produces images with better visual quality.