← back to Andy Zeng's homepage

Andy's Code Collection


A growing collection of open-source code that I've written over the years. Includes a diverse set of functions that are useful and non-trivial to implement for fast-prototyping in computer vision and robotics research.

Table of Contents

  •  Volumetric TSDF Fusion from Multiple Depth Maps

Looking for code from my publications? Check out my publications list here and follow the links.
If you have any questions, feel free to contact me by email: andyz[at]princeton[dot]edu

Volumetric TSDF Fusion of Multiple Depth Maps

CUDA/C++ code to fuse multiple registered depth maps into a projective truncated signed distance function (TSDF) voxel volume, which can then be used to create high quality 3D surface meshes and point clouds.

Requirements:
  •  NVIDA GPU with CUDA support
  •  OpenCV (tested with OpenCV 2.4.11)

Download:
See Github repository here. The demo fuses 50 registered depth maps from a directory into a projective TSDF voxel volume, and creates a 3D surface point cloud and mesh.

Seen in:
  •  3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions (CVPR 2017)
  •  Semantic Scene Completion from a Single Depth Image (CVPR 2017)
  •  Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Images (CVPR 2016)

References:
  •  A Volumetric Method for Building Complex Models from Range Images (SIGGRAPH 1996)
  •  KinectFusion: Real-Time Dense Surface Mapping and Tracking (ISMAR 2011)
  •  Scene Coordinate Regression Forests for Camera Relocalization in RGB-D Images (CVPR 2013)