Abstract
The use of LiDAR point clouds for accurate three-dimensional perception is crucial for realizing high-level autonomous driving systems. Upon considering the drawbacks of the current point cloud object-detection algorithms, this paper proposes HCNet, an algorithm that combines an attention mechanism with adaptive adjustment, starting from feature fusion and overcoming the sparse and uneven distribution of point clouds. Inspired by the basic idea of an attention mechanism, a feature-fusion structure HC module with height attention and channel attention, weighted in parallel, is proposed to perform feature-fusion on multiple pseudo images. The use of several weighting mechanisms enhances the ability of feature-information expression. Additionally, we designed an adaptively adjusted detection head that also overcomes the sparsity of the point cloud from the perspective of original information fusion. It reduces the interference caused by the uneven distribution of the point cloud from the perspective of adaptive adjustment. The results show that our HCNet has better accuracy than other one-stage-network or even two-stage-network RCNNs under some evaluation detection metrics. Additionally, it has a detection rate of 30FPS. Especially for hard samples, the algorithm in this paper has better detection performance than many existing algorithms.
Highlights
In the 21st century, automatic driving has gradually broken through the limitation of hardware, which has sped up its research process
The proposed HC module: This avoids the loss of information caused by quantifying point cloud data while enhancing the ability to express features; A self-adjusting detection head: This overcomes the impact of the sparseness and uneven distribution of the point cloud on the object detection task to a certain extent; The proposed algorithm has an inference speed of 30 fps and is comparable to the performance of the most advanced methods
The attention mechanism [30] in a neural network is a mechanism that allows the network to learn to focus on key information and ignore irrelevant information
Summary
In the 21st century, automatic driving has gradually broken through the limitation of hardware, which has sped up its research process. In order to meet the needs of three-dimensional information for target detection, new technologies represented by LiDAR have made major breakthroughs in the real-time acquisition of multi-grade three-dimensional spatial targets in recent years. This system can partially block through the woods to directly obtain high-precision three-dimensional information on the real surface, which cannot be replaced by traditional photogrammetry. A self-adjusting detection head: This overcomes the impact of the sparseness and uneven distribution of the point cloud on the object detection task to a certain extent; The proposed algorithm has an inference speed of 30 fps and is comparable to the performance of the most advanced methods
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
