English

AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentation

Computer Vision and Pattern Recognition 2020-12-10 v1 Machine Learning Robotics

Abstract

In this paper, we present an Assertion-based Multi-View Fusion network (AMVNet) for LiDAR semantic segmentation which aggregates the semantic features of individual projection-based networks using late fusion. Given class scores from different projection-based networks, we perform assertion-guided point sampling on score disagreements and pass a set of point-level features for each sampled point to a simple point head which refines the predictions. This modular-and-hierarchical late fusion approach provides the flexibility of having two independent networks with a minor overhead from a light-weight network. Such approaches are desirable for robotic systems, e.g. autonomous vehicles, for which the computational and memory resources are often limited. Extensive experiments show that AMVNet achieves state-of-the-art results in both the SemanticKITTI and nuScenes benchmark datasets and that our approach outperforms the baseline method of combining the class scores of the projection-based networks.

Keywords

Cite

@article{arxiv.2012.04934,
  title  = {AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentation},
  author = {Venice Erin Liong and Thi Ngoc Tho Nguyen and Sergi Widjaja and Dhananjai Sharma and Zhuang Jie Chong},
  journal= {arXiv preprint arXiv:2012.04934},
  year   = {2020}
}

Comments

10 pages, 7 figures, 5 tables

R2 v1 2026-06-23T20:50:21.596Z