Virtual Multi-view Fusion for 3D Semantic Segmentation
This work addresses 3D scene understanding for applications like robotics or AR/VR, but it is incremental as it builds on classic multi-view representations with improved techniques.
The paper tackles 3D semantic segmentation of meshes by proposing a virtual multi-view fusion method that selects and renders virtual views to train a 2D segmentation model, achieving significantly better results than prior multi-view approaches and competitive performance with 3D convolution methods on the ScanNet benchmark.
Semantic segmentation of 3D meshes is an important problem for 3D scene understanding. In this paper we revisit the classic multiview representation of 3D meshes and study several techniques that make them effective for 3D semantic segmentation of meshes. Given a 3D mesh reconstructed from RGBD sensors, our method effectively chooses different virtual views of the 3D mesh and renders multiple 2D channels for training an effective 2D semantic segmentation model. Features from multiple per view predictions are finally fused on 3D mesh vertices to predict mesh semantic segmentation labels. Using the large scale indoor 3D semantic segmentation benchmark of ScanNet, we show that our virtual views enable more effective training of 2D semantic segmentation networks than previous multiview approaches. When the 2D per pixel predictions are aggregated on 3D surfaces, our virtual multiview fusion method is able to achieve significantly better 3D semantic segmentation results compared to all prior multiview approaches and competitive with recent 3D convolution approaches.