Hi everyone,
I proposed and implemented a stereo matching algorithm, and right now I am trying to publish the results.
I tested the accuracy of the algorithm using quarter-size four Middlebury benchmark images ( Tsukuba, Venus, Teddy, Cones). The website is no longer accepting these images for evaluation, so I have written a simple Matlab script that computes the percentage of bad pixels in the disparity map with a threshold of 1 pixel. when I try to compare my work with the literature, my question is, how does the Middlebury website used to evaluate the sequence in this link
https://vision.middlebury.edu/stereo/eval/ ?
Salem Taki. I had something like this before. My advice is ,if you can, to contact directly the organisers of the website to help you .
Salam Achraf,
I have just found what I was looking for in the website. Thanks for the advice anyway.
edit: I thought I found it but I'm confused again.
you can visit the same thing using Matlab code scripts code
Matlab code is used to solve the problem your
you can visit the matlab cenral
Adnan Majeed Thank you for your reply, but I am really not following what you are trying to say.
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping
- Maps