Featured
CODEBASES
CodeFormer 
A blind face restoration algorithm for enhancing old photos and fixing AI arts.
VIEW MORE
Upscale-A-Video
A pratical text-guided diffusion model for enhancing real-world and AIGC videos.
VIEW MORE
BasicVSR Series
A mainstream recurrent network series for video super-resolution. [won 🥇x3 🥈x1]
BasicVSR++ RealBasicVSR
ObjectClear
A robust model for complete object-effect removal with background consistency.
VIEW MORE
GaussianAnything
A 3D diffusion model that generates high-quality and editable 3D Gaussians.
VIEW MOREFeatured
DATASETS
Stereo-Blur [CVPR'19]
Stereo-Blur is a large-scale (20,637 blur/sharp pairs) and diverse-scene (135 scenarios) dataset for stereo deblurring.
VIEW MORE
LOL-Blur [ECCV'22]
LOL-Blur contains 12,000 low-blur/normal-sharp pairs with diverse darkness and motion blurs in different scenarios.
VIEW MORE
Flare7K [NeurIPS'22]
Flare7K offers 5,000 scattering flare images and 2,000 reflective flare images for nighttime flare removal.
VIEW MORE
BracketFlare [CVPR'23]
BracketFlare contains diverse and realistic reflective flare patterns for smartphone flare (''ghosting effect'') removal.
VIEW MORE
PaintBucket-Character [CVPR'24]
PaintBucket-Character dataset includes rendered line arts alongside their colorized counterparts, featuring 22 various 3D characters.
VIEW MORE |
 |