Habitat Challenge 2019

Overview

Habitat Challenge is an autonomous navigation challenge that aims to benchmark and accelerate progress in embodied AI. In its first iteration, Habitat Challenge 2019 is based on the PointGoal task defined in Anderson et. al. We will have two tracks for the PointGoal task:

  1. RGB track: input modality for agent is RGB image.
  2. RGBD track: input modalities for agent are RGB image and Depth.

Results

RGB Leaderboard

Rank Team SPL
1 Arnold 0.805
1 Mid-level-Features 0.800
3 CHROMA 0.712
4 ARF-RL 0.699
5 MTank 0.260
6 Policy-Police 0.260

RGB-D Leaderboard

Rank Team SPL
1 Arnold 0.948
2 Pansy 0.927
3 titardrew 0.868
4 Hiccup 0.846
5 CHROMA 0.843
6 Mid-level-Features 0.815
7 ARF-RL 0.788
8 mkk1 0.738
9 cnmooc 0.693
10 hela-ppo-baseline 0.569

Prizes

The winning team for RGB and RGBD track will receive Google Cloud Platform (GCP) credits worth $10k each. Thank you for the generosity GCP!

Dates

Challenge starts April 3, 2019
Challenge submission deadline May 18, 2019

Participation

For details on how to participate, submit and train agents refer to github.com/habitat-challenge repository.

Please note that the latest submission to the test-challenge split will be used for final evaluation.