Shape Models
- Surface Correspondence Benchmark (400 watertight meshes in 20 categories is a cool starting point for playing with triangular meshes)
- AIM@Shape (Digital Shape WorkBench v5.0)
- COSEG (co-segmentation benchmark)
- ShapeNet
- Skeletonization Benchmark
- A Large Dataset of Object Scans: more than ten thousand RGBD scans of real objects & reconstructed 3D models
Shape Tools
- SAP Visual Enterprise Author 7 (Deep Exploration 6.5): browsing, converting & simplifing meshes
- MeshLab :: mesh viewer & large amount of basic operations
- Geomagic Studio: point cloud processing, reconstruction, ...
- KeyShot 5: for cool rendering used for high quality paper
- Mapple processing and visualizing point clouds and polygonal models. the use of modern GPU features for rendering and interacting with huge point clouds and polygonal models; a nice and friendly user interface for rendering (e.g., copy/restore camera, walkthrough).
- PcdViewer selection of points by sketching a rectangle, or by using a lasso tool. (source code avaliable).
- CloudCompare: 3D point cloud and mesh processing software Open Source Project.
Common Tools
- Atom: A hackable text editor
- PicPick
Shape Library
- OpenGL: Refer to references and resources of our Computer Graphics course
- libigl: a simple header-only library for geometry processingt. Combines the rapid prototyping familiar to Matlab with the performance and versatility of C++. Examples: computation of differential quantities and operators, real-time deformation, parametrization, numerical optimization and remeshing. Good tutorial!
- OpenMesh + LibQGLView: Flexible, efficient, lightweight, well documented and explained in homework of our digital geometry course.
- CGAL + LibQGLView: heavyweight, cover various areas needing geometric computation, well documented
- PCL (Point Cloud Library+VTK)
- ReconstructMe: a powerful 3D real-time scanning & reconstruction system
- Vega FEM Library: physical deformation by Jernej Barbic who also has some other interesting code
- Starlab: efficient but less documents which lead to a steep learning curve.
- ShapeOp:a header-only C++ library for static and dynamic geometry processing, using a unified framework for optimization under constraints. Examples of its applications include architectural design subject to geometric constraints, and physics simulation of elastic materials. No tutorial!
- MEPP (CGAL + QT + LibQGLView) in French
- openMVG (open Multiple View Geometry library. Basis for 3D computer vision and Structure from Motion.)
- Graphite ( + numerical solvers)
- Open Scene Graph (good renderer)
Jupyter
run code in browser or cloud free, no c++ still (you have to setup your own linux server with Cling)
- Anaconda: Leading Open Data Science Platform Powered by Python. [easy to install; with IPython notebook, Numpy, deep learning, GPU, ...]
- Sagemath cloud: for Python, Octave, R, Latex
- Binder: Turn a GitHub repo into a collection of interactive notebooks powered by Jupyter and Kubernetes.
- wakari: web-based Python data analysis
- Microsoft Azure Machine Learning: Python, R
- IBM DataScientist Workbench
Source code
- Code for computer vision, machine learning and neural computation
- patch matching via DaisyFilterFlow_CVPR2014