microsoft / microsoft/ai-dev-gallery
[SAMPLE] Super resolution alternative using open-source model
Nobody has claimed this yet.
- Dominant language
- C#
- Stars
- 1.5k
- Forks
- 223
- PR merge metrics
- No merged PRs in 30d
Description
Is your feature request related to a problem? Please describe.
We have some users who may want to use a super resolution control in their apps and we only offer a solution for this using WCR APIs. Implementing a sample that uses an open-source model would cover those users.
Describe the solution you'd like
Implement the ESRGAN model using the QNN Execution Provider. This will work on Snapdragon NPUs, DML and CPU
Describe alternatives you've considered
The only alternative is to have our users figure this process out on their own.
Additional context
Qualcomm has examples showing use for all their models in Python. Use this to help translating logic to C#. For this model, I would also use the python demo to compare outputs once you have created the C# implementation.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the linked Python ESRGAN demo from Qualcomm AI Hub and review how the existing C# samples configure the QNN Execution Provider. Implement an ESRGAN sample for Snapdragon NPUs, DirectML, and CPU, then compare its outputs with the Python demo as the completion check.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- csharp
- Domain
- ai, computer-vision, machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100