PixRestore: Unified Image Restoration via Pixel Diffusion Transformer
2026-08-17 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors introduced PixRestore, a new model that fixes low-quality images with various problems using one system. Unlike previous approaches that use parts of pretrained text-to-image models, PixRestore works directly on pixel data and is trained from scratch, which helps keep fine details. It also learns which parts of its internal data to trust more, improving the restoration process. Tests show PixRestore is efficient, accurate, and works well on different types of image damage compared to other methods.
unified image restorationlatent diffusion modelsvariational autoencoder (VAE)Diffusion Transformer (DiT)flow matchingDINO feature similarityadversarial trainingimage degradationsingle-step inferencepixel-space modeling
Authors
Lingchen Sun, Rongyuan Wu, Xiangtao Kong, Jixin Zhao, Qiaosi Yi, Yujing Sun, Shuaizheng Liu, Zhengqiang Zhang, Lei Zhang
Abstract
Unified image restoration (UIR) aims to recover high-quality (HQ) content from low-quality (LQ) images with different degradations using a single model. Most recent methods adapt large pretrained text-to-image (T2I) latent diffusion models for their strong capacity and generative priors. However, the variational autoencoder (VAE) in latent T2I models may discard restoration-sensitive details, while the open-ended synthesis prior can introduce content-inconsistent artifacts. We present PixRestore, a VAE-free pixel-space Diffusion Transformer (DiT) for UIR, where the diffusion backbone is trained entirely from scratch, without relying on T2I pretraining. PixRestore performs flow matching directly on patchified pixels, preserving fine-grained details while keeping the token sequence tractable. To adapt to different degradations, PixRestore learns to predict the reliability of layer features using LQ--HQ DINO feature similarity. Features from more reliable layers are fused as dense conditioning, while less reliable layers receive stronger HQ-feature supervision to encourage degradation removal. We train PixRestore on a large-scale corpus of diverse scenes and degradations, and further finetune it into a one-step generator using DINO-based adversarial objectives for efficient inference. Experiments on public benchmarks and real-world test sets show that, with only about 50M parameters and single-step inference, PixRestore achieves the best overall fidelity, perceptual quality, and robustness to degradations among competing UIR models while being far more efficient. Larger PixRestore variants can further boost performance, demonstrating the scalability of our pixel-space design. Code and the curated benchmark can be found at https://github.com/csslc/PixRestore.