Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

PointConvFormer: Revenge of the Point-based Convolution

About

We introduce PointConvFormer, a novel building block for point cloud based deep network architectures. Inspired by generalization theory, PointConvFormer combines ideas from point convolution, where filter weights are only based on relative position, and Transformers which utilize feature-based attention. In PointConvFormer, attention computed from feature difference between points in the neighborhood is used to modify the convolutional weights at each point. Hence, we preserved the invariances from point convolution, whereas attention helps to select relevant points in the neighborhood for convolution. PointConvFormer is suitable for multiple tasks that require details at the point level, such as segmentation and scene flow estimation tasks. We experiment on both tasks with multiple datasets including ScanNet, SemanticKitti, FlyingThings3D and KITTI. Our results show that PointConvFormer offers a better accuracy-speed tradeoff than classic convolutions, regular transformers, and voxelized sparse convolution approaches. Visualizations show that PointConvFormer performs similarly to convolution on flat areas, whereas the neighborhood selection effect is stronger on object boundaries, showing that it has got the best of both worlds.

Wenxuan Wu, Li Fuxin, Qi Shan• 2022

Related benchmarks

TaskDatasetResultRank
Semantic segmentationScanNet V2 (val)
mIoU74.5
288
Semantic segmentationScanNet v2 (test)
mIoU74.9
248
Semantic segmentationSemanticKITTI (val)
mIoU67.1
117
Scene FlowFlyingThings3D (test)
EPE3D0.0416
13
Scene Flow EstimationKITTI 2015 (test)
EPE3D (m)0.0479
7
Showing 5 of 5 rows

Other info

Follow for update