MEDIRL: Predicting the Visual Attention of Drivers via Deep Inverse Reinforcement Learning

Sonia Baee, Erfan Pakdamanian, Inki Kim, Lu Feng, Vicente Ordonez, Laura Barnes

To - Do

Revise the code and add the missed functions
Change the format of the data to the .json file (EyeCar)

Overview

Inspired by human visual attention, we introduce a Maximum Entropy Deep Inverse Reinforcement Learning (MEDIRL) framework for modeling the visual attention allocation of drivers in imminent rear-end collisions. MEDIRL is composed of visual, driving, and attention modules. Given a front-view driving video and corresponding eye fixations from humans, the visual and driving modules extract generic and driving-specific visual features, respectively. Finally, the attention module learns the intrinsic task-sensitive reward functions induced by eye fixation policies recorded from attentive drivers. MEDIRL uses the learned policies to predict visual attention allocation of drivers. We also introduce EyeCar, a new driver visual attention dataset during accident-prone situations. We conduct comprehensive experiments and show that MEDIRL outperforms previous state-of-the-art methods on driving task-related visual attention allocation on the following large-scale driving attention benchmark datasets: DR(eye)VE, BDD-A, and DADA-2000. The code and dataset are provided for reproducibility.

Dataset - EyeCar

We select 21 front-view videos that were captured in various traffic, weather, and day light conditions. Each video is 30sec in length and contains typical driving tasks (e.g., lane-keeping, merging-in, and braking) ending to rear-end collisions. Note that all the conditions were counterbalanced among all the participants. Moreover, EyeCar provides information about the speed and GPS of the ego-vehicle. In addition, each video frame comprises 4.6 vehicles on average, making EyeCar driving scenes more complex than other visual attention datasets. The EyeCar dataset contains 3.5h of gaze behavior (aggregated and raw) from the 20 participants, as well as more than 315,000 rear-end collisions video frames. In EyeCar dataset, we account for the sequence of eye fixations, and thus we emphasize on attention shift to the salient regions in a complex driving scene. EyeCar also provides a rich set of annotations (e.g., scene tagging, object bounding, lane marking, etc.). You can accesss to the data in this EyeCar. Compared to prior datasets, EyeCar is the only dataset captured from a point-of-view (POV) perspective, involving collisions, and including metadata for both speedand GPS. EyeCar also has the largest average number of vehicles per scene, and gaze data for 20 participants.

Modules of MEDIRL

The visual module: The visual module extracts low and mid-level visual cues that are useful for a variety of visual attention tasks. We rely on pre-existing models for semantic and instance segmentation, as well as depth estimation. In addition, we propose an approach to detect brake lights and traffic lights.

2. The driving module: The driving module extracts driving-specific visual features for driving tasks.

3. The attention module: Drivers pay attention to the task-related regions of the scene to filter out irrelevant information and ultimately make optimal decisions. Drivers do this with a sequence of eye fixations. To learn this process in various driving tasks ending in rear-end collisions, we cast it as a maximum inverse reinforcement learning approach.

Some Results:

Requirements

The simplest way to clone the development environment is with the anaconda-project and conda-env systems; these can be installed with Anaconda with the command

conda install anaconda anaconda-project

All necessary packages are found in anaconda-project.yml. To create an Anaconda virtual environment, run

rm environment.yml #if `environment.yml` already exists
anaconda-project export-env-spec -n medirl ./environment.yml
conda-env create -n medirl -f environment.yml
conda activate medirl

To deactivate the virtual environment, run

conda deactivate

To run a jupyter notebook inside this virtual environment, activate the virtual env medirl and run

(medirl)$ python -m ipykernel install --user --name=medirl

where the (medirl) prefix indicates that you are currently in the medirl virtual environment. Then, select the "medirl" kernel after creating or opening a jupyter notebook.

Model

You can run the main script to get the result of each component:

python main.py

We saved the best/last of models of each module in Models folder.

Data

You can accesss to the data in this EyeCar

Citing

If you find this paper/code useful, please consider citing our paper:

  @inproceedings{baee2019eyecar,
    title={EyeCar: Modeling the Visual Attention Allocation of Drivers in Semi-Autonomous Vehicles},
    author={Baee, Sonia and Pakdamanian, Erfan and Roman, Vicente Ordonez and Kim, Inki and Feng, Lu and Barnes, Laura},
    booktitle={arXiv preprint arXiv:1912.07773},
    year={2019}
  }

varunjammula/MEDIRL-EyeCar