Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/rohitgajawada/Where-are-they-looking-PyTorch

Where are they looking? - Gaze Following via Attention modelling and Deep Learning
https://github.com/rohitgajawada/Where-are-they-looking-PyTorch

Last synced: about 1 month ago
JSON representation

Where are they looking? - Gaze Following via Attention modelling and Deep Learning

Host: GitHub
URL: https://github.com/rohitgajawada/Where-are-they-looking-PyTorch
Owner: rohitgajawada
License: mit
Created: 2017-10-18T18:22:30.000Z (about 7 years ago)
Default Branch: master
Last Pushed: 2019-06-16T07:42:47.000Z (over 5 years ago)
Last Synced: 2024-08-01T19:46:11.341Z (4 months ago)
Language: Python
Homepage:
Size: 13.8 MB
Stars: 34
Watchers: 6
Forks: 10
Open Issues: 5
Metadata Files:
- Readme: README.md
- License: LICENSE

Awesome Lists containing this project

awesome-cv - Where are they looking? PyTorch Implementation
awesome-cv - Where are they looking? PyTorch Implementation

README

# Where are they looking?

This repository contains an implementation of the "Where are they looking?" paper by A. Recasens*, A. Khosla*, C. Vondrick and A. Torralba.

## Introduction

A deep neural network-based approach for gaze-following automated using a SSD face detector.

## Installation

- [Pytorch 1.1 & Torchvision](https://pytorch.org/)

## Usage

* First, download pretrained Places365 AlexNet model: https://urlzs.com/ytKK3

* Then run: python3 main.py --data_dir=`location to gazefollow dataset` --placesmodelpath=`location to places365 alexnet model`

* Please check out opts.py for other parameter changing.

## Contact

Please do get in touch with us by email for any questions, comments, suggestions you have!

* [email protected]
* [email protected]

## References

* sfzhang15's SFD detector is used for face detection (https://github.com/sfzhang15/SFD).
* Link to the NIPS 2015 paper from MIT: http://people.csail.mit.edu/khosla/papers/nips2015_recasens.pdf. Please cite them if you decide to use this project for your research.