TurkerGaze - GitLite

KO | EN

gitlite — search

#javascript #hacktoberfest #machine-learning #deep-learning #docker #macos #nodejs #cli #kubernetes #pytorch #security #ios #openai #open-source

TurkerGaze

(★ 111)

a webcam-based eye tracking game for collecting large-scale eye tracking data via crowdourcing

파일 탐색기

최종 버전 다운로드 (.zip)

- - img0.jpg
  - img1.jpg
  - img10.jpg
  - img11.jpg
  - img12.jpg
  - img13.jpg
  - img14.jpg
  - img15.jpg
  - img2.jpg
  - img3.jpg
  - img4.jpg
  - img5.jpg
  - img6.jpg
  - img7.jpg
  - img8.jpg
  - img9.jpg
- imglist.json
- system_overview.png
- - ab1.png
  - ab2.png
  - ab3.png
  - ab4.png
  - ab5.png
  - ab6.png
  - ab7.png
  - ab8.png
  - ab9.png
  - cage.png
  - pig1.png
  - pig1_gray.png
  - pig2.png
  - pig2_gray.png
  - pig3.png
  - pig3_gray.png
  - pig4.png
  - pig4_gray.png
  - pig5.png
  - pig5_gray.png
  - pig6.png
  - pig6_gray.png
- - 2.png
  - 3.png
  - 4.png
  - 5_1.png
  - 5_2.png
  - step1.png
  - step2.png
  - step3.png
  - step4.png
  - step5.png
  - whac_a_mole_gaze_px.png
- - laser_ace.mp3
  - laser_ace.ogg
  - s1.mp3
  - s1.ogg
  - s3.mp3
  - s3.ogg
  - s4.mp3
  - s4.ogg
- - 1.png
  - 10.png
  - 11.png
  - 12.png
  - 13.png
  - 2.png
  - 3.png
  - 4.png
  - 5.png
  - 6.png
  - 7.png
  - 8.png
  - 9.png
  - blink.png
  - bright.png
  - dark.png
- - extractHistFeature.js
  - jsfeat.js
  - OpticalFlow.js
- - bowser.min.js
  - howler.min.js
  - jquery-1.11.0.min.js
  - jquery.zclip.js
  - json.js
  - snap.svg-min.js
  - ZeroClipboard.swf
- - MLModel.js
  - RidgeRegression.js
- monitor.js
- parameter.js
- setup.js
- utils.js
- install.sh
- index.js
- - digital.ttf
  - lcd.ttf
  - zorque.ttf
- - pu_logo.png
- font.css
- main.css
.gitignore
bundle.js
index.html
LICENSE
package.json
README.md
visualizer.html
webpack.config.js

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# TurkerGaze
TurkerGaze: Crowdsourcing Saliency with Webcam based Eye Tracking

### Instructions
```bash
> ./scripts/install.sh
> npm install
> npm start

```

Created by Pingmei Xu, Jianxiong Xiao at Princeton Vision Group.

### Introduction
TurkerGaze is a webcam-based eye tracking game for collecting large-scale eye tracking data via crowdourcing. It is implemented in javascript and the details were described in an [arXiv tech report](http://arxiv.org/abs/1504.06755).

![system overview](http://isun.cs.princeton.edu/TurkerGaze/demo/system_overview.png)

### Citing
If you find TurkerGaze useful in your research, please consider citing:

@article{xu15arXiv,
        Author = {Pingmei Xu, Krista A Ehinger, Yinda Zhang, Adam Finkelstein, Sanjeev R. Kulkarni, Jianxiong Xiao},
        Title = {Rich feature hierarchies for accurate object detection and semantic segmentation},
        Booktitle = {arXiv:1504.06755},
        Year = {2015}
    }

### Usage

0. See a demo
	0. Setup a local web server, download the folder, open 'pugazetrackr.html' to run the eye tracking task. Save the result data to a local file and visualize the result by 'visualizer.html'.
	0. You can also try it here:
	[eye tracking task](http://isun.cs.princeton.edu/TurkerGaze/pugazetrackr.html)
	[visualization](http://isun.cs.princeton.edu/TurkerGaze/visualizer.html)

0. User your own images
	0. Create a .json object with two fields: 'gaze' and 'memory' like './demo/imglist.json'. 'gaze' contains the images that you want to present for free-viewing, and 'memory' contains images for the memory test.
	0. Pass the path of this .json file by url parameter 'imglist'. For example, http://isun.cs.princeton.edu/TurkerGaze/pugazetrackr.html?imglist=your_imglist_url
	0. Run the task!

### License
TurkerGaze is released under the MIT License (refer to the LICENSE file for details).