SaptakBhoumik/TinyVision

TinyVision is an evolving project focused on designing ultra-lightweight image classification models with minimal parameter counts. The goal is to explore what’s actually necessary for fundamental vision tasks by combining handcrafted feature preprocessing with highly efficient CNN architectures.

22
/ 100
Experimental

This project helps machine learning researchers and engineers explore how to build very small, efficient image classification models. It takes in image datasets, like those for classifying cats vs. dogs or CIFAR-10, and outputs highly compact neural network models that can still achieve good accuracy. The primary users are those focused on minimizing model size and computational resources for computer vision tasks.

No commits in the last 6 months.

Use this if you are a machine learning researcher or engineer interested in designing and experimenting with ultra-lightweight image classification models for resource-constrained environments.

Not ideal if you are looking for a ready-to-use, production-grade image classification solution or a library for general-purpose computer vision tasks beyond lightweight model experimentation.

deep-learning-research model-compression efficient-AI image-classification computer-vision-engineering
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

13

Forks

Language

Jupyter Notebook

License

CC-BY-SA-4.0

Last pushed

Aug 06, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/SaptakBhoumik/TinyVision"

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