Multi-Task Attention Network (MTAN)

This repository contains the source code of Multi-Task Attention Network (MTAN) and baselines from the paper, End-to-End Multi-Task Learning with Attention, introduced by Shikun Liu, Edward Johns, and Andrew Davison.

Experiments

Image-to-Image Predictions (One-to-Many)

Under the folder im2im_pred, we have provided our proposed network alongside with all the baselines on NYUv2 dataset presented in the paper. All the models were written in pytorch. So please first make sure you have pytorch 1.0 framework or above installed in your machine.

Download our pre-processed NYUv2 dataset here which we used in the paper. The original NYUv2 dataset can be found in here with pre-computed ground-truth normals from here.

All the models (files) are built with SegNet and described in the following table:

File Name	Type	Flags	Comments
model_segnet_single.py	Single	task, dataroot	standard single task learning
model_segnet_stan.py	Single	task, dataroot	our approach whilst applied on one task
model_segnet_split.py	Multi	weight, dataroot, temp, type	multi-task learning baseline in which the shared network splits at the last layer (also known as hard-parameter sharing)
model_segnet_dense.py	Multi	weight, dataroot, temp	multi-task learning baseline in which each task has its own paramter space (also known as soft-paramter sharing)
model_segnet_cross.py	Multi	weight, dataroot, temp	our implementation of the Cross Stitch Network
model_segnet_mtan.py	Multi	weight, dataroot, temp	our approach

For each flag, it represents

Flag Name	Usage	Comments
task	pick one task to train: semantic (semantic segmentation, depth-wise cross-entropy loss), depth (depth estimation, l1 norm loss) or normal (normal prediction, cos-similarity loss)	only availiable in single-task learning
dataroot	directory root for NYUv2 dataset	just put under the folder im2im_pred to avoid any concerns :D
weight	weighting options for multi-task learning: equal (direct summation of all task losses), DWA (our proposal), uncert (our implementation of the Weight Uncertainty Method)	only available in multi-task learning
temp	hyper-parameter temperature in DWA weighting option	to determine the softness of task weighting
type	different versions of multi-task baseline split: standard, deep, wide	only available in the baseline split

To run any model, cd im2im_pred/ and run python MODEL_NAME.py --FLAG_NAME 'FLAG_OPTION'.

Visual Decathlon Challenge (Many-to-Many)

We have also provided source code for the recently proposed Visual Decathlon Challenge for which we build MTAN based on Wide Residual Network from the implementation here.

To run the code, first download the dataset and devkit at the official Visual Decathlon Challenge website here and put it in the folder visual_decathlon. Then, put decathlon_mean_std.pickle into the folder of the downloaded dataset decathlon-1.0-data.

Finally, run python model_wrn_mtan.py for training python model_wrn_eval.py --dataset 'imagenet' and 'notimagenet' for evaluation and python coco_results.py for COCO format for online evaluation.

Other Notices

1. The provided code is highly optimised for readibility. If you find any unusual behaviour, please post an issue or directly contact my email below.
2. Training the provided code will result slightly better performances than the reported numbers in the paper for image-to-image prediction tasks (the rankings stay the same). If you want to compare any models in the paper for image-to-image prediction tasks, please re-run the model directly with your own training strategies (learning rate, optimiser, etc) and keep all training strategies consistent to ensure fairness. To compare results in Visual Decathlon Challenge, you may directly check out the results in the paper. To compare with your own research, please build your multi-task network with the same backbone architecture: SegNet for image-to-image tasks, and Wide Residual Network for the Visual Decathlon Challenge.
3. From my personal experience, designing a better architecture is usually more helpful (and easier) than finding a better task weighting in multi-task learning.

Citation

If you found this code/work to be useful in your own research, please considering citing the following:

@article{liu2018mtan,
  title={End-to-end multi-task learning with attention},
  author={Liu, Shikun and Johns, Edward and Davison, Andrew J},
  journal={arXiv preprint arXiv:1803.10704},
  year={2018}
}

Contact

If you have any questions, please contact [email protected].