Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Metric: A-DISTS, locally adaptive DISTS #346

Open
luisbarrancos opened this issue Feb 18, 2023 · 0 comments
Open

Metric: A-DISTS, locally adaptive DISTS #346

luisbarrancos opened this issue Feb 18, 2023 · 0 comments
Labels
feature New feature or request

Comments

@luisbarrancos
Copy link

luisbarrancos commented Feb 18, 2023

Is your feature request related to a problem? Please describe.
DISTS is a good metric for global structure, yet natural images have variations in local structure. Taking these into consideration would provide more meaningful metrics for these types of images.

Describe the solution you'd like
A pytorch implementation or extension of DISTS to take into account local structure. Adding this together with IGSTQA and perhaps STSIM would provide the users with a meaningful toolbox for structural perceptual similarity, both at global and local level.

Describe alternatives you've considered
DISTS and LPIPS only.

Additional context
The original paper Locally Adaptive Structure and Texture Similarity for Image Quality Assessment by Ding et al, was presented in 2021, and can be found here (PDF, Arxiv) but there's no reference implementation, to the best of my knowledge.

Update: there is a reference implementation here, https://github.com/dingkeyan93/A-DISTS/

@luisbarrancos luisbarrancos added the feature New feature or request label Feb 18, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature New feature or request
1 participant