Efficient Semantic Scene Completion Network with Spatial Group Convolution

Abstract

We introduce Spatial Group Convolution (SGC) for accelerating the computation of 3D dense prediction tasks. SGC is orthogonal to group convolution, which works on spatial dimensions rather than feature channel dimension. It divides input voxels into different groups, then conducts 3D sparse convolution on these separated groups. As only valid voxels are considered when performing convolution, computation can be significantly reduced with a slight loss of accuracy. The proposed operations are validated on semantic scene completion task, which aims to predict a complete 3D volume with semantic labels from a single depth image. With SGC, we further present an efficient 3D sparse convolutional network, which harnesses a multiscale architecture and a coarse-to-fine prediction strategy. Evaluations are conducted on the SUNCG dataset, achieving state-of-the-art performance and fast speed. Code is available at https://github.com/zjhthu/SGC-Release.git

An oral paper in ECCV 2018, and the code is available at https://github.com/zjhthu/SGC-Release.git

References in corpus (6)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Compressing Deep Convolutional Networks using Vector Quantization
• Submanifold Sparse Convolutional Networks
• Matterport3D: Learning from RGB-D Data in Indoor Environments
• Large-Scale 3D Shape Reconstruction and Segmentation from ShapeNet Core55
• 3D Point Cloud Classification and Segmentation using 3D Modified Fisher Vector Representation for Convolutional Neural Networks