Name: Yuzhi ZHAO
Type: User
Company: City University of Hong Kong
Bio: Ph.D., CityU EE.
B.Eng, HUST EIC (电信卓越班)
Location: Hong Kong, China
Blog: https://zhaoyuzhi.github.io/
Yuzhi ZHAO's Projects
TensorFlow implementation of a CNN based mechanical science paper
The face detectors for human, cat, dog using dlib and opencv
Automatic crop videos and produce static, interpolation, and averaged frames or new videos for making blur dataset
百度图片爬虫,基于python3
pure Python implement of BM3D
Some functions for demosaicking camera RAW files including .raw / .NEF format
ChildPredictor: A Child Face Prediction Framework with Disentangled Learning. IEEE TMM, 2022
Contrastive Language-Image Pretraining
Image to prompt with BLIP and CLIP, compatible with Stable Diffusion
Modified context encoder PyTorch implementation
Let us control diffusion models!
吴恩达老师的机器学习课程个人笔记
D2HNet: Joint Denoising and Deblurring with Hierarchical Network for Robust Night Image Restoration. ECCV, 2022
PyTorch DDPM implementation
PyTorch implementation of CVPR 2018 paper: DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks
Provide pre-trained MobileNetv123 and MobileNet-DSC FPN models for DeblurGANv2
TensorFlow implementation of cascade CNN, trained level-by-level
CVPR19 - Deep Network Interpolation for Continuous Imagery Effect Transition
Code for NeurIPS 2020 paper: Blind Video Temporal Consistency via Deep Video Prior
The PyTorch implementation of ICCV 2019 oral paper: free-form inpainting (deepfillv2), especially Gated Conv
When you want to make a dataset for deep learning, it may help you
PyTorch implementation of TIP 2017 paper DnCNN, while real-time validation and visualization
Implementation of our sRGB denoising method in CVPR 2020 NTIRE Workshop
English and Chinese translation using the pre-trained models
Face manipulation using dlib
Learning Blind Video Temporal Consistency (ECCV 2018)
Python Script to download hundreds of images from 'Google Images'. It is a ready-to-run code!
CVPR NTIRE 2020 Spectral Reconstruction Challenge 1st Place Paper