cs.CVApr 14, 2026

HyperLiDAR: Adaptive Post-Deployment LiDAR Segmentation via Hyperdimensional Computing

Authors: Ivannia Gomez Moreno, Yi Yao, Ye Tian, Xiaofan Yu, Flavio Ponzina, Michael Sullivan, Jingyi Zhang, Mingyu Yang, 

Organizations: UCSD · UC Merced · SDSU · UMich

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

LiDAR semantic segmentation plays a pivotal role in 3D scene understanding for edge applications such as autonomous driving. However, significant challenges remain for real-world deployments, particularly for on-device post-deployment adaptation. Real-world environments can shift as the system navigates through different locations, leading to substantial performance degradation without effective and timely model adaptation. Furthermore, edge systems operate under strict computational and energy constraints, making it infeasible to adapt conventional segmentation models (based on large neural networks) directly on-device. To address the above challenges, we introduce HyperLiDAR, the first lightweight, post-deployment LiDAR segmentation framework based on Hyperdimensional Computing (HDC). The design of HyperLiDAR fully leverages the fast learning and high efficiency of HDC, inspired by how the human brain processes information. To further improve the adaptation efficiency, we identify the high data volume per scan as a key bottleneck and introduce a buffer selection strategy that focuses learning on the most informative points. We conduct extensive evaluations on two state-of-the-art LiDAR segmentation benchmarks and two representative devices. Our results show that HyperLiDAR outperforms or achieves comparable adaptation performance to state-of-the-art segmentation methods, while achieving up to a 13.8x speedup in retraining.

Figures & tables

Figure 1. The pipeline of HyperLiDAR including three stages. The pre-deployment training is performed in the cloud. The post-deployment adaptation and inference are performed on edge devices, taking streaming LiDAR scans as inputs.
DatasetModelmIoU (%)IoU (%)
carbicyclemotorcycletruckpedestrianterrain
SemanticKITTI ( Behley et al., 2019 )CENET ( Cheng et al., 2022 )56.994.932.333.875.453.375.9
SALUDA ( Michele et al., 2024 )49.193.31.030.516.940.090.3
HyperLiDAR (Ours)54.095.221.450.761.959.666.7
HyperLiDAR-full (Ours)55.795.239.852.962.260.968.9
nuScenes ( Caesar et al., 2020 )CENET ( Cheng et al., 2022 )54.586.812.857.960.749.061.0
Table 1. Performance on SemanticKITTI and NuScenes.
Figure 3
More in this topic

3D Semantic Segmentation

3D semantic segmentation assigns meaning to points, voxels, meshes, or reconstructed scene elements for spatial understanding.

See all 137 papers

Explore similar work

CardsList
Jun 29, 2026

No Adaptation Without Observation: Observability-Constrained Test-Time Prompt Tuning for LiDAR Semantic Segmentation

LiDAR semantic segmentation often degrades under real-world deployment due to evolving sensing conditions, while collecting new annotations for retraining is impractical. Test-time adaptation (TTA) updates model parameters online using pseudo-label supervision, but directly applying standard TTA strategies to LiDAR data is challenging. Because pseudo-label reliability is spatially heteroscedastic under range-dependent sparsity and occlusion, uniform updates on globally shared parameters can inject unstable gradients and destabilize adaptation. We propose a geometry-constrained test-time prompt tuning framework for LiDAR semantic segmentation. Our method estimates per-location sensing reliability from depth-consistent beam terminations and neighborhood support, and uses it to reweight spatial supervision. Adaptation is confined to lightweight prompt adapters inserted into a frozen backbone, with spatial gating to prevent unreliable regions from perturbing globally shared representations. A temporally smoothed prototype alignment strategy further stabilizes online updates by accumulating reliable semantic evidence over time. Experiments on standard LiDAR benchmarks demonstrate improved adaptation stability and segmentation performance under deployment variations without additional annotations.

Linlian Jiang, Wentao Ju, Sadman Rakib Pinon +4

Concordia University · Mila – Quebec AI Institute · University of Toronto

Test-Time Adaptation3D Semantic SegmentationTest-Time Prompt Tuning
AskPDF
Mar 28, 2026

TerraSeg: Self-Supervised Ground Segmentation for Any LiDAR

LiDAR perception is fundamental to robotics, enabling machines to understand their environment in 3D. A crucial task for LiDAR-based scene understanding and navigation is ground segmentation. However, existing methods are either handcrafted for specific sensor configurations or rely on costly per-point manual labels, severely limiting their generalization and scalability. To overcome this, we introduce TerraSeg, the first self-supervised, domain-agnostic model for LiDAR ground segmentation. We train TerraSeg on OmniLiDAR, a unified large-scale dataset that aggregates and standardizes data from 12 major public benchmarks. Spanning almost 22 million raw scans across 15 distinct sensor models, OmniLiDAR provides unprecedented diversity for learning a highly generalizable ground model. To supervise training without human annotations, we propose PseudoLabeler, a novel module that generates high-quality ground and non-ground labels through self-supervised per-scan runtime optimization. Extensive evaluations demonstrate that, despite using no manual labels, TerraSeg achieves state-of-the-art results on nuScenes, SemanticKITTI, and Waymo Perception while delivering real-time performance. Our code and model weights are publicly available.

Ted Lentsch, Santiago Montiel-Marín, Holger Caesar +1

Department of Cognitive Robotics, Delft University of Technology · Department of Electronics, University of Alcalá

3D Point Cloud SegmentationSelf-Supervised Learning
AskPDF
May 20, 2026

LiteViLNet: Lightweight Vision-LiDAR Fusion Network for Efficient Road Segmentation

Road segmentation is a fundamental perception task for autonomous driving and mobile robotics, where both appearance and geometric cues must be processed under edge-computing constraints. Existing multi-modal approaches often improve accuracy with large encoders or expensive global interaction, which limits their use on embedded platforms. We present LiteViLNet, a lightweight RGB-geometry fusion network that combines a MobileNetV3 RGB encoder with a 0.12M-parameter depth-wise-separable geometry encoder. A multi-scale feature fusion module performs modality-specific enhancement, global-query cross-modal interaction, and adaptive gating, while a depth-wise large-kernel bridge enlarges the contextual support of the deepest representation with low overhead. The resulting U-Net-style decoder uses deep supervision only during training. On the KITTI Road benchmark, the 14.04M-parameter full model obtains 97.23±0.15%97.23\pm0.15\% MaxF. On the held-out ORFD test set under the released OFF-Net evaluation protocol, the full model achieves 96.74±0.09%96.74\pm0.09\% F-score and 93.68±0.18%93.68\pm0.18\% IoU. On a Jetson Orin NX, model-only PyTorch FP16 inference reaches 22.18±0.2122.18\pm0.21 FPS; a separate TensorRT FP16 measurement reaches 68.73±0.0668.73\pm0.06 FPS on the Jetson. Camera-depth adaptations and perception-and-control demonstrations on three heterogeneous robot platforms further illustrate the portability of the dual-stream design.

Daojie Peng, Bingtao Wang, Fulong Ma +2

The Hong Kong University of Science and Technology (Guangzhou) · The Shandong University

Edge InferenceCamera-LiDAR FusionAutonomous DrivingAutonomous Driving Perception
AskPDF