3 papers
cs.CV2026
HiTokSR: A Coarse-to-Fine Tokenizer with Hierarchical Codebooks for High-Fidelity Real-World Image Super-Resolution
Mingxi Li
Vector-quantized (VQ) generative models have shown promising results in real-world image super-resolution (Real-ISR). However, existing methods typically rely on a monolithic laten…
cs.CV2026
Spectrum Matching: a Unified Perspective for Superior Diffusability in Latent Diffusion
Mang Ning, Mingxiao Li, Le Zhang +4
In this paper, we study the diffusability (learnability) of variational autoencoders (VAE) in latent diffusion. First, we show that pixel-space diffusion trained with an MSE object…
cs.CV2025
DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT Space
Mang Ning, Mingxiao Li, Jianlin Su +6
This paper explores image modeling from the frequency space and introduces DCTdiff, an end-to-end diffusion generative paradigm that efficiently models images in the discrete cosin…