3 papers
cs.CV2026
Disentangled Textual Priors for Diffusion-based Image Super-Resolution
Lei Jiang, Xin Liu, Xinze Tong +4
Image Super-Resolution (SR) aims to reconstruct high-resolution images from degraded low-resolution inputs. While diffusion-based SR methods offer powerful generative capabilities,…
cs.CV2026
GLAD: Generative Language-Assisted Visual Tracking for Low-Semantic Templates
Xingyu Luo, Yidong Cai, Jie Liu +3
Vision-language tracking has gained increasing attention in many scenarios. This task simultaneously deals with visual and linguistic information to localize objects in videos. Des…
cs.CV2025
Pure-Pass: Fine-Grained, Adaptive Masking for Dynamic Token-Mixing Routing in Lightweight Image Super-Resolution
Junyu Wu, Jie Liu, Jie Tang +1
Image Super-Resolution (SR) aims to reconstruct high-resolution images from low-resolution counterparts, but the computational complexity of deep learning-based methods often hinde…