4D-StOP: Panoptic Segmentation of 4D LiDAR using Spatio-temporal Object Proposal Generation and Aggregation
It addresses the problem of accurate 4D scene understanding for autonomous driving, offering a significant performance gain over existing approaches.
The paper tackles 4D panoptic LiDAR segmentation by introducing 4D-StOP, which uses voting-based spatio-temporal proposals and geometric feature aggregation, achieving a state-of-the-art score of 63.9 LSTQ on SemanticKITTI, a 7% improvement over previous methods.
In this work, we present a new paradigm, called 4D-StOP, to tackle the task of 4D Panoptic LiDAR Segmentation. 4D-StOP first generates spatio-temporal proposals using voting-based center predictions, where each point in the 4D volume votes for a corresponding center. These tracklet proposals are further aggregated using learned geometric features. The tracklet aggregation method effectively generates a video-level 4D scene representation over the entire space-time volume. This is in contrast to existing end-to-end trainable state-of-the-art approaches which use spatio-temporal embeddings that are represented by Gaussian probability distributions. Our voting-based tracklet generation method followed by geometric feature-based aggregation generates significantly improved panoptic LiDAR segmentation quality when compared to modeling the entire 4D volume using Gaussian probability distributions. 4D-StOP achieves a new state-of-the-art when applied to the SemanticKITTI test dataset with a score of 63.9 LSTQ, which is a large (+7%) improvement compared to current best-performing end-to-end trainable methods. The code and pre-trained models are available at: https://github.com/LarsKreuzberg/4D-StOP.