Online Segmentation of LiDAR Sequences: Dataset and Algorithm

16 Jun 2022  ·  Romain Loiseau, Mathieu Aubry, Loïc Landrieu ·

Roof-mounted spinning LiDAR sensors are widely used by autonomous vehicles. However, most semantic datasets and algorithms used for LiDAR sequence segmentation operate on $360^\circ$ frames, causing an acquisition latency incompatible with real-time applications. To address this issue, we first introduce HelixNet, a $10$ billion point dataset with fine-grained labels, timestamps, and sensor rotation information necessary to accurately assess the real-time readiness of segmentation algorithms. Second, we propose Helix4D, a compact and efficient spatio-temporal transformer architecture specifically designed for rotating LiDAR sequences. Helix4D operates on acquisition slices corresponding to a fraction of a full sensor rotation, significantly reducing the total latency. Helix4D reaches accuracy on par with the best segmentation algorithms on HelixNet and SemanticKITTI with a reduction of over $5\times$ in terms of latency and $50\times$ in model size. The code and data are available at: https://romainloiseau.fr/helixnet

PDF Abstract

Datasets


Introduced in the Paper:

HelixNet

Used in the Paper:

SemanticKITTI
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Real-Time Semantic Segmentation HelixNet Helix4D mIoU (1/5 rotation) 78.7 # 1
Inference Time (ms) (1/5 rotation) 19 # 1

Methods


No methods listed for this paper. Add relevant methods here