Alexander Clegg

This is hosted as a gist https://colab.research.google.com/gist/mathfac/3c15af549cc4498a0b0f0b84130a7203/eccv2020_habitat_challenge_tutorial.ipynb

Robot API moved to Habitat-lab.

Note we have stopped supporting colab and converted tutorials to local Jupyter notebooks instead.

To follow up here, the mesh data is exposed in C++, but you'd need to write accessors and bindings to expose it as in #1149. Note that internally, there are...

Hey @pfmephisto, very odd issue here. I'm not sure why the depth image is empty. It may need to be normalized to the clipping range (e.g. 10 m). The semantic...

@vaibhavoutat we used YCB from https://huggingface.co/datasets/ai-habitat/ycb as the clutter and pick-able objects for rearrangement which were inserted into HSSD scenes. The configs for generating episodes show which objects we used....

Hey @alre5639, The dataset you are modifying here is actually a PointNavDataset which contains PointNavEpisodes. These are produced by the generator: https://github.com/facebookresearch/habitat-lab/tree/main/habitat-lab/habitat/datasets/pointnav. This generator is fed a configuration file which...

CUDA 11 should be fine for training with Habitat on the current and past versions. 11.8 is preferred for PyTorch, but we use 11.3 on CI for unit testing without...

Thanks for raising this, @sanjayharesh, I can reproduce this issue locally. Looks like this particular code isn't run our CI, so possible it has become stale.

Hey @BaiLing09 Most likely this is an issue with the coordinate system configuration for the scene meshes. In MP3D, the semantic mesh is a separate asset instantiating a separate SceneGraph....