HiMo: High-Speed Objects Motion Compensation in Point Clouds

Abstract

LiDAR point cloud is essential for autonomous vehicles, but motion distortions from dynamic objects degrade the data quality. While previous work has considered distortions caused by ego motion, distortions caused by other moving objects remain largely overlooked, leading to errors in object shape and position. This distortion is particularly pronounced in high-speed environments such as highways and in multi-LiDAR configurations, a common setup for heavy vehicles. To address this challenge, we introduce HiMo, a pipeline that repurposes scene flow estimation for non-ego motion compensation, correcting the representation of dynamic objects in point clouds. During the development of HiMo, we observed that existing self-supervised scene flow estimators often produce degenerate or inconsistent estimates under high-speed distortion. We further propose SeFlow++, a real-time scene flow estimator that achieves state-of-the-art performance on both scene flow and motion compensation. Since well-established motion distortion metrics are absent in the literature, we introduce two evaluation metrics: compensation accuracy at a point level and shape similarity of objects. We validate HiMo through extensive experiments on Argoverse 2, ZOD and a newly collected real-world dataset featuring highway driving and multi-LiDAR-equipped heavy vehicles. Our findings show that HiMo improves the geometric consistency and visual fidelity of dynamic objects in LiDAR point clouds, benefiting downstream tasks such as semantic segmentation and 3D detection.

Figure 1: Multi-LiDARs are equipped in our heavy vehicles to avoid self-occlusion. (a) shows an example placement with 6 LiDARs. The point colors in (b-c) correspond to the LiDAR from which the points are captured. (b) illustrates the distortion of static structure due to fast-moving ego vehicle. Raw shows the raw data, w. egc shows the ego-motion compensation results. (c) demonstrates distortion caused by motion of other objects, which depends on the velocity of the said objects. In such case, ego-motion compensation (w. ego-motion comp.) alone is insufficient. In comparison, our HiMu pipeline (w. HiMo motion comp.) successfully undistorted the point clouds completely, resulting in an accurate representation of the objects.

Results

Here are supplementary video and full qualitative results for the main paper. Please refer to our main paper for more details analysis.

Compensation result on different dataset using our HiMo pipeline

In this section, we present three interactive comparison between raw data with ego-motion compensation only and our HiMo pipeline using SeFlow++. Results include our Scania highway data, Argoverse 2 data and Zenseact Open Dataset (ZOD) that cover three different configurations and sensor types.

Data scene id: batch_184_20211217173504, timestamp: 075

Data scene id: 76916359-96f4-3274-81fe-bb145d497c11, timestamp: 315968469059874000

Data scene id: 000018, timestamp: 16547760125175530

Comparison on different scene flow methods inside HiMo pipeline

We show the undistorted effect of different flow methods inside our HiMo pipeline as a supplement to the main paper for clarity.

Data scene id: batch_062_20211022162703, timestamp: 078
The raw data with ego-motion compensation still shows a truck object exhibiting significant distortion. It provides the full qualitative results on the performance of different flow methods inside HiMo pipeline.

---

Data scene id: batch_184_20211217173504, timestamp: 075
The raw data with ego-motion compensation still shows big point distortion of two fast moving car objects in the scenarios. It provides the full qualitative results on the performance of different flow methods inside HiMo pipeline.

BibTeX

If you find our work useful in your research, please consider citing:

Copy

@article{zhang2025himp,
    title={HiMo: High-Speed Objects Motion Compensation in Point Clouds},
    author={Zhang, Qingwen and Khoche, Ajinkya and Yang, Yi and Ling, Li and Sina, Sharif Mansouri and Andersson, Olov and Jensfelt, Patric},
    year={2025},
    journal={arXiv preprint arXiv:2503.00803},
}

---