ybsong00 / drht Goto Github PK
View Code? Open in Web Editor NEWImage Correction via Deep Reciprocating HDR Transformation
Image Correction via Deep Reciprocating HDR Transformation
Hi developer, thank you first that shared the code of paper Image Correction via Deep Reciprocating HDR Transformation . Now I want to run your code on my own computer, but your introduce page didn't tell the detail of the version of the libraries that code used, especially the version tensorflow and opencv.
If this request don't bother you, could you tell me the detail of the running environment?
where can i get it
We used ldr2hdr.npz parameters from the drive link of the pre-trained model. The problem we are facing is that the image we use as input comes out as the output without any changes. The model does not seem to be doing any computation on the image at all.
Also, in the folder structure you have specified, "input --- input ldr images" and "samples --- ldr results", I see the opposite. The LDR jpg images (sky is blown out) are in the samples folder whereas the HDR jpg images (clouds visible) is in the input folder. Am I wrong?
I would like to know more about the training images you used to build your pre-trained model. If I can have that dataset, that will be highly appreciated. I am working on a similar project and I am using this paper as a baseline. My project is limited till the work done in "HDR image reconstruction from a single exposure using deep CNNs"
Hi, I carefully read the paper and found you generate the ground truth LDR images of city scene panorama dataset as follow:
we use the black-box Adobe Photoshop software to empirically generate ground truth LDR images with human supervision.
So I was wondering is that process done automatically, otherwise whether you will release the corresponding data?
Thank you~
Hello, I have downloaded hdrcnn model and put it in ~/DRHT/checkpoint/. After that, I tried to run ldr2hdr_test.py, however, error occurred and said, "TypeError: init() got an unexpected keyword argument 'layer'". I don't know how to solve this problem.
The environment was ubuntu16.04+python3.5+tensorflow-gpu1.10.0.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.