implemented a simple AI-model integration pipeline #14
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #11
What was changed?
Introduced a client-side AI model integration that lets users upload an image and receive relevant search results based on visual similarity. The application now runs a Hugging Face model entirely in the browser, computes image embeddings, and compares them to a small set of reference images to generate results. (or at least this is the goal)
Why was it changed?
We needed a generic and easily maintainable way to integrate AI models into the app without relying on a separate backend. This ensures that we can swap in different models or pipelines with minimal effort, while meeting the requirement of handling images on the client and returning relevant search results.
How was it changed?
A local inference pipeline was added that downloads and runs a pre-trained model (CLIP) in the user’s browser. When an image is uploaded, it will be converted into an embedding, normalized, and compared to precomputed embeddings of reference images to determine similarity. This approach keeps the logic entirely on the client side, removing any dependency on external servers or APIs for model inference.
NOTE this implementation might change before merging