https://github.com/cloud-cv/grad-cam

:rainbow: :camera: Gradient-weighted Class Activation Mapping (Grad-CAM) Demo
https://github.com/cloud-cv/grad-cam

cnn deep-learning demo grad-cam machine-learning torch

Last synced: 17 days ago
JSON representation

:rainbow: :camera: Gradient-weighted Class Activation Mapping (Grad-CAM) Demo

• Host: GitHub
• URL: https://github.com/cloud-cv/grad-cam
• Owner: Cloud-CV
• Created: 2016-07-15T09:20:25.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2018-08-13T18:49:48.000Z (over 6 years ago)
• Last Synced: 2024-11-09T10:41:42.639Z (2 months ago)
• Topics: cnn, deep-learning, demo, grad-cam, machine-learning, torch
• Language: HTML
• Homepage: http://gradcam.cloudcv.org/
• Size: 5.72 MB
• Stars: 109
• Watchers: 8
• Forks: 26
• Open Issues: 8
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        

# Grad-CAM: Gradient-weighted Class Activation Mapping

[![Join the chat at https://gitter.im/Cloud-CV/Grad-CAM](https://badges.gitter.im/Cloud-CV/Grad-CAM.svg)](https://gitter.im/Cloud-CV/Grad-CAM?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)

Grad-CAM uses the class-specific gradient information flowing into the final convolutional layer of a CNN to produce a coarse localization map of the important regions in the image. It is a novel technique for making CNN more 'transparent' by producing **visual explanations** i.e visualizations showing what evidence in the image supports a prediction. You can play with Grad-CAM demonstrations at the following links:

**Arxiv Paper Link**: https://arxiv.org/abs/1610.02391

### Grad-CAM VQA Demo: http://gradcam.cloudcv.org/vqa

![Imgur](http://i.imgur.com/6jB4lAq.gif)

### Grad-CAM Classification Demo: http://gradcam.cloudcv.org/classification

![Imgur](http://i.imgur.com/a1IiQg4.gif)

### Grad-CAM Captioning Demo: http://gradcam.cloudcv.org/captioning

![Imgur](http://i.imgur.com/BsOOpIn.gif)

## Installing / Getting started

We use RabbitMQ to queue the submitted jobs. Also, we use Redis as backend for realtime communication using websockets.

All the instructions for setting Grad-CAM from scratch can be found  [here](https://github.com/Cloud-CV/Grad-CAM/blob/master/INSTALLATION.md)

Note: For best results, its recommended to run the Grad-CAM demo on GPU enabled machines.

## Interested in Contributing?

Cloud-CV always welcomes new contributors to learn the new cutting edge technologies. If you'd like to contribute, please fork the repository and use a feature branch. Pull requests are warmly welcome.

if you have more questions about the project, then you can talk to us on our [Gitter Channel](https://gitter.im/Cloud-CV/Grad-CAM).  

## Acknowledgements

- [VQA_LSTM_CNN](https://github.com/VT-vision-lab/VQA_LSTM_CNN)

- [HieCoAttenVQA](https://github.com/jiasenlu/HieCoAttenVQA)

- [NeuralTalk2](https://github.com/karpathy/neuraltalk2/)

- [PyTorch](https://github.com/hughperkins/pytorch)