刘国友's repositories
SFHQ-dataset
Synthetic Faces High Quality (SFHQ) Dataset. 425,000 curated 1024x1024 synthetic face images
WMatrixMedian
[SIGGRAPH Asia 2022 Technical Papers' Best Paper Award]
insightface
State-of-the-art 2D and 3D Face Analysis Project
AdelaiDepth
This repo contains the projects: 'Virtual Normal', 'DiverseDepth', and '3D Scene Shape'. They aim to solve the monocular depth estimation, 3D scene reconstruction from single image problems.
DAMO-YOLO
DAMO-YOLO: a fast and accurate object detection method with some new techs, including NAS backbones, efficient RepGFPN, ZeroHead, AlignedOTA, and distillation enhancement.
PGMPI
The implementation of Real-Time Neural Character Rendering with Pose-Guided Multiplane Images
mugen
A command-line music video generator based on rhythm
WCT2
Software that can perform photorealistic style transfer without the need of any post-processing steps.
BasicTAD
BasicTAD: an Astounding RGB-Only Baselinefor Temporal Action Detection
RepGhost
RepGhost: A Hardware-Efficient Ghost Module via Re-parameterization
The-Change-You-Want-to-See
The official implementation of the paper The Change You Want to See (WACV 2023).
SpecTNT-pytorch
Unofficial implementation of SpecTNT in pytorch
Learning-to-Incorporate-Texture-Saliency-Adaptive-Attention-to-Image-Cartoonization
Code of paper "Learning to Incorporate Texture Saliency Adaptive Attention to Image Cartoonization", ICML 2022.
PhotoWCT2
PhotoWCT2: Compact Autoencoder for Photorealistic Style Transfer Resulting from Blockwise Training and Skip Connections of High-Frequency Residuals
polyblur
An unofficial Python (Numpy and Pytorch) implementation of "Polyblur: Removing mild blur by polynomial reblurring" by Mauricio Delbracio and collaborators.
CondensedMovies
Story-Based Retrieval with Contextual Embeddings. Largest freely available movie video dataset. [ACCV'20]
VToonify
[SIGGRAPH Asia 2022] VToonify: Controllable High-Resolution Portrait Video Style Transfer
rg-cache
Code for the paper "Reflectance-guided, contrast-accumulated histogram equalization" published in ICASSP 2020.
architectural_style_transfer
Code and data release for ICCP 2022 paper "Time-of-Day Neural Style Transfer for Architectural Photographs".
CDFI
[CVPR 2021] CDFI: Compression-Driven Network Design for Frame Interpolation
thecartoonist
Cartoon filter (thick lines, cel shading). Implementation of "Image Abstraction by Structure Adaptive Filtering" by Jan Eric Kyprianidis and Jürgen Döllner