Point-Voxel CNN for Efficient 3D Deep Learning

TL;DR

PVCNN combines point-based and voxel-based methods to create a memory- and computation-efficient 3D deep learning model that outperforms existing approaches in accuracy and speed.

Contribution

The paper introduces PVCNN, a novel hybrid architecture that reduces memory and computation costs while improving accuracy in 3D deep learning tasks.

Findings

Achieves 10x GPU memory reduction over voxel-based models.

Outperforms state-of-the-art point-based models with 7x speedup.

Outperforms PointNet with 2x speedup and higher accuracy.

Abstract

We present Point-Voxel CNN (PVCNN) for efficient, fast 3D deep learning. Previous work processes 3D data using either voxel-based or point-based NN models. However, both approaches are computationally inefficient. The computation cost and memory footprints of the voxel-based models grow cubically with the input resolution, making it memory-prohibitive to scale up the resolution. As for point-based networks, up to 80% of the time is wasted on structuring the sparse data which have rather poor memory locality, not on the actual feature extraction. In this paper, we propose PVCNN that represents the 3D input data in points to reduce the memory consumption, while performing the convolutions in voxels to reduce the irregular, sparse data access and improve the locality. Our PVCNN model is both memory and computation efficient. Evaluated on semantic and part segmentation datasets, it achieves much higher accuracy than the voxel-based baseline with 10x GPU memory reduction; it also outperforms the state-of-the-art point-based models with 7x measured speedup on average. Remarkably, the narrower version of PVCNN achieves 2x speedup over PointNet (an extremely efficient model) on part and scene segmentation benchmarks with much higher accuracy. We validate the general effectiveness of PVCNN on 3D object detection: by replacing the primitives in Frustrum PointNet with PVConv, it outperforms Frustrum PointNet++ by 2.4% mAP on average with 1.5x measured speedup and GPU memory reduction.