SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentation
Preprint
Abstract
LiDAR point-cloud segmentation is an important problem for many applications. For large-scale point cloud segmentation, the \textit{de facto} method is to project a 3D point cloud to get a 2D LiDAR image and use convolutions to process it. Despite the similarity between regular RGB and LiDAR images, we discover that the feature distribution of LiDAR images changes drastically at different image locations. Using standard convolutions to process such LiDAR images is problematic, as convolution filters pick up local features that are only active in specific regions in the image. As a result, the capacity of the network is under-utilized and the segmentation performance decreases. To fix this, we propose Spatially-Adaptive Convolution (SAC) to adopt different filters for different locations according to the input image. SAC can be computed efficiently since it can be implemented as a series of element-wise multiplications, im2col, and standard convolution. It is a general framework such that several previous methods can be seen as special cases of SAC. Using SAC, we build SqueezeSegV3 for LiDAR point-cloud segmentation and outperform all previous published methods by at least 3.7% mIoU on the SemanticKITTI benchmark with comparable inference speed.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Deformable Kernel Networks for Joint Image FilteringInternational Journal of Computer Vision, 2020
- RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds2020
- xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation2020
- Linking Points With Labels in 3D: A Review of Point Cloud Semantic SegmentationIEEE Geoscience and Remote Sensing Magazine, 2020
- Dynamic Graph CNN for Learning on Point CloudsACM Transactions on Graphics, 2019
- SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud2019
- SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloud2018
- Perceptual Losses for Real-Time Style Transfer and Super-Resolution2016
- Dark Memory and Accelerator-Rich System Optimization in the Dark Silicon EraIEEE Design & Test, 2016
- Microsoft COCO: Common Objects in Context2014