Rarity-Aware Discrete Diffusion with Spatially Consistent Decoding for Photo-Realistic Image Super-Resolution
2026-07-20 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors study a new way to improve photo-realistic image super-resolution using discrete diffusion models, which treat images as groups of visual tokens that can be predicted in parallel. They identify two challenges: rare but important textures are often missed, and parallel prediction can cause small artifacts. To fix this, their method, DiMOO-SR, focuses more on rare tokens during training and uses a neighborhood check during inference to keep image parts consistent. Their tests show that DiMOO-SR produces high-quality images efficiently, suggesting discrete diffusion is promising for this task.
super-resolutiondiffusion modelsdiscrete diffusionvisual tokensautoregressive modelsinverse frequency samplingspatial consistencyparallel decodingimage reconstruction
Authors
Ao Li, Yapeng Du, Yi Xin, Lei Zhu, Le Zhang, Guangtao Zhai, Ce Zhu, Xiaohong Liu
Abstract
Continuous diffusion models have become the dominant paradigm for photo-realistic image Super-Resolution (SR), but they typically formulate reconstruction as continuous signal-level denoising and incorporate semantic priors through external conditioning modules. This makes it less direct to exploit the unified token-based scaling paradigm of modern multimodal models. Autoregressive models provide a more native semantic representation by modeling images as discrete visual tokens, yet their causal decoding is inefficient for high-resolution reconstruction. Discrete diffusion offers a promising middle ground by enabling non-causal, parallel prediction over visual tokens. However, directly adapting discrete diffusion to SR remains non-trivial due to two task-specific challenges: (1) the long-tailed distribution of visual tokens, which under-represents rare but perceptually critical textures; and (2) spatially inconsistent parallel decoding, which may introduce isolated artifacts. To address these issues, we propose DiMOO-SR, a rarity-aware multimodal discrete diffusion framework for photo-realistic image SR. During training, Inverse Frequency Sampling (IFS) prioritizes under-represented but information-rich tokens. During inference, Spatial Consistency Ranking (SCR) refines token confidence using local neighborhood agreement to improve structural coherence. Extensive experiments on widely used real-world SR benchmarks demonstrate that DiMOO-SR achieves competitive perceptual quality with only a few parallel decoding steps, highlighting the potential of discrete diffusion for generative image super-resolution. The code will be released upon publication.