(AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Network
arXiv:2102.04530
Abstract
Autonomous robotic systems and self driving cars rely on accurate perception of their surroundings as the safety of the passengers and pedestrians is the top priority. Semantic segmentation is one the essential components of environmental perception that provides semantic information of the scene. Recently, several methods have been introduced for 3D LiDAR semantic segmentation. While, they can lead to improved performance, they are either afflicted by high computational complexity, therefore are inefficient, or lack fine details of smaller instances. To alleviate this problem, we propose AF2-S3Net, an end-to-end encoder-decoder CNN network for 3D LiDAR semantic segmentation. We present a novel multi-branch attentive feature fusion module in the encoder and a unique adaptive feature selection module with feature map re-weighting in the decoder. Our AF2-S3Net fuses the voxel based learning and point-based learning into a single framework to effectively process the large 3D scene. Our experimental results show that the proposed method outperforms the state-of-the-art approaches on the large-scale SemanticKITTI benchmark, ranking 1st on the competitive public leaderboard competition upon publication.
arXiv comment10 pages, 6 figures, 4 tables
References in corpus (4)
- YOLOv3: An Incremental Improvement
- Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentation
- KPRNet: Improving projection-based LiDAR semantic segmentation
- The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks
Cited by in corpus (5)
- Gaussian Radar Transformer for Semantic Segmentation in Noisy Radar Data
- DRINet++: Efficient Voxel-as-point Point Cloud Segmentation
- FusionPainting: Multimodal Fusion with Adaptive Attention for 3D Object Detection
- PolarStream: Streaming Lidar Object Detection and Segmentation with Polar Pillars
- VIN: Voxel-based Implicit Network for Joint 3D Object Detection and Segmentation for Lidars