rhysdg/vision-at-a-clip

Low-latency ONNX and TensorRT based zero-shot classification and detection with contrastive language-image pre-training based prompts

19
/ 100
Experimental

This tool helps non-technical users quickly analyze and understand what's in their images using natural language. You input images and text descriptions (like "a photo of a dog" or "spaceman"), and it tells you what objects are present or the likelihood of different descriptions matching the image. It's designed for anyone needing fast visual search, classification, or object detection without complex machine learning setup.

No commits in the last 6 months.

Use this if you need to rapidly identify objects, classify images, or search through visual content using simple text prompts, especially when you need high performance and low latency.

Not ideal if you require highly specialized object recognition that is not described by everyday language or if your images contain extremely fine-grained, niche details that require extensive pre-training on specific datasets.

image-analysis visual-search content-moderation object-detection media-asset-management
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 3 / 25

How are scores calculated?

Stars

44

Forks

1

Language

Jupyter Notebook

License

Last pushed

Aug 31, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/rhysdg/vision-at-a-clip"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.