https://github.com/StanfordVL/taskonomy

Taskonomy: Disentangling Task Transfer Learning [Best Paper, CVPR2018]
https://github.com/StanfordVL/taskonomy

Last synced: about 2 months ago
JSON representation

Taskonomy: Disentangling Task Transfer Learning [Best Paper, CVPR2018]

• Host: GitHub
• URL: https://github.com/StanfordVL/taskonomy
• Owner: StanfordVL
• License: mit
• Created: 2017-12-18T20:08:09.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2022-12-08T01:00:20.000Z (almost 2 years ago)
• Last Synced: 2024-07-05T04:05:32.375Z (3 months ago)
• Language: Python
• Homepage: https://taskonomy.vision
• Size: 26.9 MB
• Stars: 834
• Watchers: 33
• Forks: 146
• Open Issues: 32
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-Multi-Task-Learning - Taskonomy

README

        
# [Taskonomy: Disentangling Task Transfer Learning](https://taskonomy.vision/)

This repository contains:

- [pretrained models (task bank)](https://github.com/StanfordVL/taskonomy/tree/master/taskbank) [PyTorch + TensorFlow].

- [dataset](https://github.com/StanfordVL/taskonomy/tree/master/data)

- [reference code](https://github.com/StanfordVL/taskonomy/tree/master/code)

- [task affinity analyses and results](https://github.com/StanfordVL/taskonomy/tree/master/results)

for the the following paper:

#### [Taskonomy: Disentangling Task Transfer Learning](https://taskonomy.vision/) (CVPR 2018, Best Paper Award)

Amir Zamir, Alexander Sax*, William Shen*, Leonidas Guibas, Jitendra Malik, Silvio Savarese. 

| [TASK BANK](https://github.com/StanfordVL/taskonomy/tree/master/taskbank) | [DATASET](https://github.com/StanfordVL/taskonomy/tree/master/data) |

|:-----|:-----|

| The `taskbank` folder contains information about our pretrained models, and scripts to download them. There are sample outputs, and links to live demos. | The `data` folder contains information and statistics about the dataset, some sample data, and [instructions for how to download the full dataset](https://docs.omnidata.vision/starter_dataset_download.html#Examples). |

| [![models](taskbank/assets/web_assets/taskbank_tile_GIF_crop.gif)](https://github.com/StanfordVL/taskonomy/tree/master/taskbank) | [![cauthron](data/assets/cauthron_small.png)](https://github.com/StanfordVL/taskonomy/tree/master/data) |

| [Task affinity analyses and results](https://github.com/StanfordVL/taskonomy/tree/master/results)| [Website](http://taskonomy.vision/) |

|:----|:----|

| This folder contains the raw and normalized data used for measuring task affinities. | The webpage of the project with links to assets and demos. |

|[![task affinity analyses and results](data/assets/affinity.jpg)](https://github.com/StanfordVL/taskonomy/tree/master/results) | [![Website front page](data/assets/web_frontpage_small.png)](http://taskonomy.vision/)|

 

 

## Citation

If you find the code, models, or data useful, please cite this paper:

```

@inproceedings{zamir2018taskonomy,

  title={Taskonomy: Disentangling Task Transfer Learning},

  author={Zamir, Amir R and Sax, Alexander and and Shen, William B and Guibas, Leonidas and Malik, Jitendra and Savarese, Silvio},

  booktitle={2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},

  year={2018},

  organization={IEEE}

}

```