# AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentation Source: [AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentation]() Collection index: [llms.txt]() Data updated: 2026-01-04T03:21:09.653Z Authors: Venice Erin Liong, Thi Ngoc Tho Nguyen, Sergi Widjaja, Dhananjai Sharma, Zhuang Jie Chong Published: 2020-12-09T09:34:25.000Z ## 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. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Semantic Segmentation on SemanticKITTI single-scan](): Semantic segmentation; SemanticKITTI single-scan; mIoU: 65.3. - [Semantic Segmentation on nuScenes 1.0 (val)](): Semantic segmentation; nuScenes 1.0 (val); mIoU: 76.1. - [3D Semantic Segmentation on nuScenes Lidar-Seg (val)](): 3D Semantic Segmentation; nuScenes Lidar-Seg (val); mIoU: 76.1. - [LiDAR Semantic Segmentation on SemanticKITTI (val)](): LiDAR Semantic Segmentation; SemanticKITTI (val); mIoU: 65.2. - [LiDAR Semantic Segmentation on nuScenes (val)](): LiDAR Semantic Segmentation; nuScenes (val); mIoU: 77.2. - [3D Semantic Segmentation on nuScenes (test)](): 3D Semantic Segmentation; nuScenes (test); mIoU: 77.27. - [Semantic Segmentation on nuScenes (test)](): Semantic segmentation; nuScenes (test); mIoU: 77.3. - [LiDAR Semantic Segmentation on SemanticKITTI (test)](): LiDAR Semantic Segmentation; SemanticKITTI (test); mIoU: 65.3. - [LiDAR Semantic Segmentation on nuScenes official (test)](): LiDAR Semantic Segmentation; nuScenes official (test); mIoU: 77.3. - [Semantic Segmentation on SemanticKITTI (test)](): Semantic segmentation; SemanticKITTI (test); mIoU: 65.3. - [Semantic Segmentation on nuScenes (val)](): Semantic segmentation; nuScenes (val); mIoU (Segmentation): 0.761. ## Original sources and implementation [Original paper]() [PDF]()