/omnisyn

OmniSyn: Synthesizing 360 Videos with Wide-baseline Panoramas

Primary LanguagePython

OmniSyn

This repo contains code for training and evaluating OmniSyn.
OmniSyn predicts depth from two panoramas and then renders meshes using PyTorch3D from intermediate positions.

Getting Started

  1. Install PyTorch and PyTorch3d
  2. Download our processed carla dataset.

Currently, the main scripts are:

  • train_depth.py which trains the depth estimator only.
  • train_inpainting.py to train the inpainting component.

Requirements

  • Install v0.2.5 of PyTorch3d
    pip install 'git+https://github.com/facebookresearch/pytorch3d.git@v0.2.5'