17 citations · 32 across the 10 of their papers we have counts for
6 papers · 1 filter
Unsupervised Real-World Super-Resolution via Rectified Flow Degradation Modelling
Hongyang Zhou, Xiaobin Zhu, Liuling Chen +4
Unsupervised real-world super-resolution (SR) faces critical challenges due to the complex, unknown degradation distributions in practical scenarios. Existing methods struggle to g…
Similarity Matters: A Novel Depth-guided Network for Image Restoration and A New Dataset
Junyi He, Liuling Chen, Hongyang Zhou +5
Image restoration has seen substantial progress in recent years. However, existing methods often neglect depth information, which hurts similarity matching, results in attention di…
VCapsBench: A Large-scale Fine-grained Benchmark for Video Caption Quality Evaluation
Shi-Xue Zhang, Hongfa Wang, Duojun Huang +3
Video captions play a crucial role in text-to-video generation tasks, as their quality directly influences the semantic coherence and visual fidelity of the generated videos. Altho…
DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Framework
Henrique Morimitsu, Xiaobin Zhu, Roberto M. Cesar +2
Optical flow estimation is essential for video processing tasks, such as restoration and action recognition. The quality of videos is constantly increasing, with current standards…
Video-Language Alignment via Spatio-Temporal Graph Transformer
Shi-Xue Zhang, Hongfa Wang, Xiaobin Zhu +5
Video-language alignment is a crucial multi-modal task that benefits various downstream applications, e.g., video-text retrieval and video question answering. Existing methods eith…
Inverse-like Antagonistic Scene Text Spotting via Reading-Order Estimation and Dynamic Sampling
Shi-Xue Zhang, Chun Yang, Xiaobin Zhu +3
Scene text spotting is a challenging task, especially for inverse-like scene text, which has complex layouts, e.g., mirrored, symmetrical, or retro-flexed. In this paper, we propos…