gabrieleilertsen/hdrcnn
HDR image reconstruction from a single exposure using deep CNNs
This tool helps photographers, videographers, and graphic designers transform standard 8-bit images or video frames into High Dynamic Range (HDR) images. It takes a single, standard exposure image where bright areas might be overexposed and fills in the lost detail, providing a richer, more detailed output image. It's designed for anyone needing to recover highlight information in their visual media.
541 stars. No commits in the last 6 months.
Use this if you need to recover blown-out highlights or details in very bright areas of a single photograph or video frame, without needing multiple exposures.
Not ideal if your input images are already HDR or if you require fine-grained manual control over the entire reconstruction process rather than an automated solution.
Stars
541
Forks
103
Language
Python
License
BSD-3-Clause
Category
Last pushed
Oct 28, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/gabrieleilertsen/hdrcnn"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
cszn/KAIR
Image Restoration Toolbox (PyTorch). Training and testing codes for DPIR, USRNet, DnCNN, FFDNet,...
INVOKERer/DeepRFT
The code for 'Intriguing Findings of Frequency Selection for Image Deblurring' and 'Deep...
emidan19/deep-tempest
Restoration for TEMPEST images using deep-learning
VinAIResearch/blur-kernel-space-exploring
Exploring Image Deblurring via Blur Kernel Space (CVPR'21)
soumik12345/MIRNet
Tensorflow implementation of MIRNet for Low-light image enhancement