GPView A GPU Accelerated Voxelization Framework for 3D CAD models.

GPView voxelizes CAD models with a hybrid resolution. Supported CAD models are Wavefront (.obj) file and .off file formats. Output is in the form of binary .raw files with contiguous array. Level1 resolution is a coarse level of voxelization and Level2 is a finer level of voxelization of particular Level1 voxels in a nested structure.

Some significant features of GPView are:

• Visualize triangulated CAD models (obj, off & stl)
• Voxelize solid models to custom resolution and aspect ratio
• Create multi-level voxelization for solid models
• Save voxelized models to binary files

Project Links

GPView is a free and open-source project hosted on GitHub.

Publications

• Gavin Young, Adarsh Krishnamurthy; GPU-accelerated generation and rendering of multi-level voxel representations of solid models, Computers & Graphics, 75:11–24, 2018.

• Sambit Ghadai, Aditya Balu, Soumik Sarkar, Adarsh Krishnamurthy; Learning localized features in 3D CAD models for manufacturability analysis of drilled holes, Computer Aided Geometric Design, 62:263–275, 2018.

• Sambit Ghadai, Aditya Balu, Xian Lee, Soumik Sarkar, Adarsh Krishnamurthy; Multi-level 3D convolutional neural network for object recognition, NVIDIA GPU Technology Conference, 2018.

• Sambit Ghadai, Aditya Balu, Adarsh Krishnamurthy, Soumik Sarkar; Learning and visualizing localized geometric features using 3D-CNN: An application to manufacturability analysis of drilled holes, NIPS Symposium on Interpretable Machine Learning, 2017.

• Aditya Balu, Sambit Ghadai, Soumik Sarkar, Adarsh Krishnamurthy; GPU-accelerated deep learning framework for cyber-enabled manufacturing, NVIDIA GPU Technology Conference, Session ID S7397, 2017.