https://github.com/mrgemy95/visual-interaction-networks-pytorch

This's an implementation of deepmind Visual Interaction Networks paper using pytorch
https://github.com/mrgemy95/visual-interaction-networks-pytorch

computer-vision convolutional-neural-networks deep-learning deepmind machine-learning nerual-network physics pytorch video-prediction-models visual-interaction-networks

Last synced: 12 days ago
JSON representation

This's an implementation of deepmind Visual Interaction Networks paper using pytorch

• Host: GitHub
• URL: https://github.com/mrgemy95/visual-interaction-networks-pytorch
• Owner: MrGemy95
• License: mit
• Created: 2017-10-13T10:50:15.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2018-02-02T15:53:29.000Z (almost 7 years ago)
• Last Synced: 2024-10-23T02:55:08.296Z (21 days ago)
• Topics: computer-vision, convolutional-neural-networks, deep-learning, deepmind, machine-learning, nerual-network, physics, pytorch, video-prediction-models, visual-interaction-networks
• Language: Python
• Homepage:
• Size: 6.01 MB
• Stars: 166
• Watchers: 5
• Forks: 25
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Visual-Interaction-Networks

An implementation of Deepmind visual interaction networks in Pytorch.

 # Introduction

> For the purpose of understanding the challenge of relational reasoning. they publised VIN that involves predicting the future in a physical scene. From just a glance, humans can infer not only what objects are where, but also what will happen to them over the upcoming seconds, minutes and even longer in some cases. For example, if you kick a football against a wall, your brain predicts what will happen when the ball hits the wall and how their movements will be affected afterwards (the ball will ricochet at a speed proportional to the kick and - in most cases - the wall will remain where it is).







## Architecture







### Data

I used [I@jaesik817](https://github.com/jaesik817/Interaction-networks_tensorflow) physics engine to generate the data.

Just run the `physics_engine.py`

## Usage

### Main Dependencies

``` 

Python 3.5

pytorch 0.3

numpy 1.13.1

```

### RUN

- Edit configration file to meet your need.

- Run `vin.py`

### References

* https://github.com/jaesik817/visual-interaction-networks_tensorflow