Multi-layered Stixels Prompted by Semantic Information
摘要
Obstacle detection and ranging are essential for intelligent vehicles, and multi-layer stixel-world is a way to represent the information of the scene in front of the vehicle. Based on the original multi-layer stixel-world algorithm, the three-category semantic segmentation model of road surface, obstacle and sky is used as the prompt information to help improve the representation effect. Experiments show that the prompting of semantic information can indeed better distinguish between roads and obstacles in the real scene.