DualTSR is presented, a unified framework that formulates STISR as coupled continuous-discrete generation and achieves the best FID, LPIPS, ACC, and NED among the compared methods at both X2 and X4.
Axi Niu, Knag Zhang, Qing-Sen Yan et al.· 0 citations
A novel Dual-Prior Collaborative Network (DPC-Net) is proposed, which achieves high-quality restoration by jointly exploiting degradation-semantic coupled priors and low-level visual priors and propagates coupled representations to the decoder.
Zhaokun He, Kangbiao Shi, Axi Niu et al.· 0 citations
TGFusion is proposed, a text-guided latent-space flow matching framework that unifies degradation suppression and cross-modal fusion and achieves superior or competitive performance in perceptual quality, image naturalness, structural-detail preservation, and infrared-saliency retention, while remaining robust across d...
Axi Niu, Jiehua Li, Kang Zhang et al.· 0 citations
This work proposes MeanSR, a one-step perceptual SR method that learns an LR-conditioned average velocity field to directly capture the finite-time transition from degraded or noisy inputs to plausible HR outputs and introduces a Stage-Aware Temporal Sampling strategy to improve trajectory learning.
Axi Niu, Jiawei Kou, Kang Zhang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.