abhshkdz/neural-vqa-attention

:question: Attention-based Visual Question Answering in Torch

38
/ 100
Emerging

This project helps computer vision researchers and AI developers build systems that can understand images and answer questions about them. It takes an image and a natural language question as input, then produces a text answer along with a 'heatmap' showing where in the image the model focused its attention to derive that answer. It's designed for those exploring visual question answering models and their interpretability.

101 stars. No commits in the last 6 months.

Use this if you need a straightforward, interpretable model to understand how a system 'looks' at an image to answer a question, rather than needing the absolute highest accuracy.

Not ideal if you require state-of-the-art accuracy for highly sensitive or critical visual question answering applications.

visual question answering image understanding explainable AI attention mechanisms computer vision research
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 21 / 25

How are scores calculated?

Stars

101

Forks

34

Language

Jupyter Notebook

License

Last pushed

Aug 13, 2017

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/abhshkdz/neural-vqa-attention"

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