Update README.md

README.md

@@ -6,19 +6,15 @@ sdk: static
 pinned: false
 ---
 
-We are now using [HuggingFace](https://huggingface.co/ShapeNet) to distribute ShapeNet data. Request access in the specific ShapeNet dataset repositories. Make sure to carefully fill out the information required in the access request form.
+We are now using [HuggingFace](https://huggingface.co/ShapeNet) to distribute ShapeNet data. *Please request access in the specific ShapeNet dataset repositories linked below, and make sure to carefully review the terms of use and fully fill out the information required in the access request form*.
 
-**ShapeNetCore** is a subset of the full ShapeNet dataset with single clean 3D models and manually verified category and alignment annotations. It covers 55 common object categories with about 51,300 unique 3D models. The 12 object categories of PASCAL 3D+, a popular computer vision 3D benchmark dataset, are all covered by ShapeNetCore. You can get an archive of the v1 release from this [LINK](https://huggingface.co/datasets/ShapeNet/ShapeNetCore-archive/blob/main/ShapeNetCore.v1.zip) (30.3GB). See [v1 README.txt](https://shapenet.org/resources/releases/shapenetcore/v1/README.txt) for details. There is also a newer v2 release of ShapeNetCore with improved quality of model geometry and fixed issues relating to materials and textures: [HERE](https://huggingface.co/datasets/ShapeNet/ShapeNetCore-archive/blob/main/ShapeNetCore.v2.zip) (~25GB). This package provides improved .obj+.mtl format model data that replaces the .obj+.mtl data in ShapeNetCore.v1, and is part of an upcoming public data release. See [v2 README.txt](https://shapenet.org/resources/releases/shapenetcore/v2/README.txt) for details.
+Quick links:
+- **ShapeNetCore**: [v1 archive](https://huggingface.co/datasets/ShapeNet/ShapeNetCore-archive/blob/main/ShapeNetCore.v1.zip) (30.3GB) | [v1 README.txt](https://shapenet.org/resources/releases/shapenetcore/v1/README.txt) | [v2 archive](https://huggingface.co/datasets/ShapeNet/ShapeNetCore-archive/blob/main/ShapeNetCore.v2.zip) (~25GB) | [v2 README.txt](https://shapenet.org/resources/releases/shapenetcore/v2/README.txt).
+- **ShapeNetSem**: [v0 archive](https://huggingface.co/datasets/ShapeNet/ShapeNetSem-archive/blob/main/ShapeNetSem.zip) (~12GB) | [v0 README.txt](https://shapenet.org/resources/releases/shapenetsem/v0/README.txt).
+- **Software tools**: see list in [tools.md](./tools.md)
+- [**Terms of Use**](./TERMS.md)
+- [**ShapeNet tech report**](http://arxiv.org/abs/1512.03012)
 
-**ShapeNetSem** is a smaller, more densely annotated subset consisting of 12,000 models spread over a broader set of 270 categories. In addition to manually verified category labels and consistent alignments, these models are annotated with real-world dimensions, estimates of their material composition at the category level, and estimates of their total volume and weight. You should be able to download all the v0 release data [HERE](https://huggingface.co/datasets/ShapeNet/ShapeNetSem-archive/blob/main/ShapeNetSem.zip) (~12GB). See [ShapeNetSem v0 README.txt](https://shapenet.org/resources/releases/shapenetsem/v0/README.txt) for details.
-
-Training, validation and test splits for ShapeNetCore are available in CSV format: [HERE](http://shapenet.cs.stanford.edu/shapenet/obj-zip/SHREC16/all.csv). For more information on the format see the evaluation details at [HERE](https://shapenet.org/shrec16)
-
-In case it would be helpful to you, we have open sourced a model viewing and rendering framework: [HERE](https://github.com/ShapeNet/shapenet-viewer).
-
-Please be mindful of the [terms of use](./TERMS.md) and especially the restrictions on redistributing this data due to copyright
-concerns.
-
-We appreciate any feedback on the dataset. If you use ShapeNet data, please cite the [main ShapeNet technical report](http://arxiv.org/abs/1512.03012) and any other relevant ShapeNet publications.
+We appreciate any feedback on the dataset. If you use ShapeNet data, please cite the above main ShapeNet technical report and any other relevant ShapeNet publications.
 
 The ShapeNet team

tools.md

@@ -0,0 +1,25 @@
+### Rendering
+- [ShapeNet Viewer](https://github.com/ShapeNet/shapenet-viewer) For viewing and rendering ShapeNet models, we provide this open-sourced model viewing framework. This Java+Scala code was used to render the ShapeNet model screenshots and thumbnails. It allows for easy batch rendering of ShapeNet models, including generating views for use with the Mitsuba renderer.
+- [Mitsuba renderer](https://github.com/shi-jian/mitsuba-shapenet) The Mitsuba raytracing framework has also been forked by one of our users for rendering ShapeNet models.
+- [Blender](https://developer.blender.org/project/view/2/) Blender is an open source 3D model creation and rendering suite. It has good support in terms of documentation and several video tutorials on youtube.
+- [SmartScenes Toolkit](https://github.com/smartscenes/sstk) SmartScenes Toolkit is a javascript, [WebGL](https://developer.mozilla.org/en-US/docs/Web/API/WebGL_API) based toolkit built on top of [three.js](https://threejs.org/) that can be used to do offscreen rendering and web-based annotation of ShapeNet models.
+
+### Deep learning
+- [PyTorch3D](https://pytorch3d.org/) PyTorch3D is a deep learning library (on top of PyTorch) for 3D data that
+supports differentiable rendering. It now supports easy loading of [ShapeNetCore](https://pytorch3d.readthedocs.io/en/latest/_modules/pytorch3d/datasets/shapenet/shapenet_core.html)
+dataset.
+
+### Mesh processing and conversion
+- [Assimp Library](https://github.com/assimp/assimp) Assimp is a portable open source library to import various 3D model formats. It also provides a full asset conversion pipleline.
+- [MeshLab](https://github.com/cnr-isti-vclab/meshlab) Provides tools for processing and editing 3D triangular meshes specifically editing, cleaning, healing, texturing and converting of meshes.
+- [PyMesh](https://github.com/qnzhou/PyMesh) PyMesh is a rapid prototyping platform focused on geometry processing. It provides a set of common mesh processing functionalities and interfaces with a number of state-of-the-art open source packages to combine their power seamlessly under a single developing environment.
+- [Python-pcl](https://github.com/strawlab/python-pcl) Python binding to the Point Cloud Library(PCL) which provides APIs for loading, segmentation, smoothing and filtering of PCD.
+- [Mayavi](https://github.com/enthought/mayavi) Provides interactive visualization of 3D data with a simple cleaning interface in Python script. It reads several data formats like VTK, PLOT3D etc.
+- [gptoolbox](https://github.com/alecjacobson/gptoolbox/) This is a toolbox of useful matlab functions for geometry processing. It has utility functions for several geometry processing functionalities, constrained optimization and image processing.
+- [Toolbox Graph](http://www.mathworks.com/matlabcentral/fileexchange/5355-toolbox-graph) This toolbox written in Matlab contain functions to create synthetic graph and compute shortest path (dijkstra and isomap algorithm).This toolbox can load triangulations from files and then display the resulting mesh. It allows to compute various laplacian operator, and the to compute parameterization using spectral decomposition, harmonic mapping, free boundary harmonic mapping, and isomap.
+- [Toolbox Fast Marching](https://www.mathworks.com/matlabcentral/mlcdownloads/downloads/submissions/6110/versions/3/previews/toolbox_fast_marching/html/content.html?access_key=) This toolbox provides with implementation of Fast Marching algorithm on various 3D representations like volumes, meshes etc.
+
+### Voxelization
+- [Binvox](http://www.patrickmin.com/binvox/) Voxelization for ShapeNet models are computed using binvox.
+
+Please also let us know if you have tools for working with ShapeNet that you would like to share with the community, and we will list them here.