Skip to content

Latest commit

 

History

History
40 lines (26 loc) · 1.98 KB

README.md

File metadata and controls

40 lines (26 loc) · 1.98 KB
Raw

Data Processing Scripts

Various scripts for (pre-)processing of data. Functionality includes rendering, identification of physically plausible object poses generation of watertight meshes from arbitrary triangle soups, identification of uncertain regions as introduced in Shape Completion with Prediction of Uncertain Regions and more.

Overview

Rendering

  • render_kinect.py: Render depth images from a given camera pose and intrinsics. Also simulates Kinect depth noise.
  • render_kinect_parallel.py: Runs render_kinect.py in parallel for multiple camera poses and intrinsics.
  • render_data.py Uses BlenderProc for the rendering.

Find physically plausible poses

The script generate_physics_poses.py generates physically plausible poses for sets of objects. It uses the Bullet Physics Engine to simulate the object falling onto a plane.

Generate watertight meshes

The script make_watertight.py generates watertight meshes from triangle soups. It is based on mesh-fusion with various improvements for quality, robustness and speed.

Usage

make_watertight path/to/ShapeNetCore.v1

Use --out_dir /path/to/output/directory to specify the output directory.

Find uncertain regions

The script find_uncertain_regions.py identifies regions with ambiguous occupancy given the current point of view. For more details refer to Shape Completion with Prediction of Uncertain Regions.

Usage

find_uncertain_regions -cn shapenet_uncertain

Add vis.split=val or vis.split=test to find uncertain regions for the validation or test set, respectively. Use log.verbose=True to get detailed information about the process and vis.show=True to visualize the results.