ICP-4D: Bridging Iterative Closest Point and LiDAR Panoptic Segmentation

• URL: http://arxiv.org/abs/2512.18991v1
• Date: Mon, 22 Dec 2025 03:13:08 GMT
• Title: ICP-4D: Bridging Iterative Closest Point and LiDAR Panoptic Segmentation
• Authors: Gyeongrok Oh, Youngdong Jang, Jonghyun Choi, Suk-Ju Kang, Guang Lin, Sangpil Kim,
• Abstract summary: ICP-4D is a training-free framework that unifies spatial and temporal reasoning through geometric relations among instance-level point sets.<n>To stabilize association under noisy instance predictions, we introduce a Sinkhorn-based soft matching.<n>Our experiments across both SemanticKITTI and panoptic nuScenes demonstrate that our method consistently outperforms state-of-the-art approaches.
• Score: 44.68614934602709
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Dominant paradigms for 4D LiDAR panoptic segmentation are usually required to train deep neural networks with large superimposed point clouds or design dedicated modules for instance association. However, these approaches perform redundant point processing and consequently become computationally expensive, yet still overlook the rich geometric priors inherently provided by raw point clouds. To this end, we introduce ICP-4D, a simple yet effective training-free framework that unifies spatial and temporal reasoning through geometric relations among instance-level point sets. Specifically, we apply the Iterative Closest Point (ICP) algorithm to directly associate temporally consistent instances by aligning the source and target point sets through the estimated transformation. To stabilize association under noisy instance predictions, we introduce a Sinkhorn-based soft matching. This exploits the underlying instance distribution to obtain accurate point-wise correspondences, resulting in robust geometric alignment. Furthermore, our carefully designed pipeline, which considers three instance types-static, dynamic, and missing-offers computational efficiency and occlusion-aware matching. Our extensive experiments across both SemanticKITTI and panoptic nuScenes demonstrate that our method consistently outperforms state-of-the-art approaches, even without additional training or extra point cloud inputs.

Related papers

• Adaptive Point-Prompt Tuning: Fine-Tuning Heterogeneous Foundation Models for 3D Point Cloud Analysis [51.37795317716487]
  We propose the Adaptive Point-Prompt Tuning (APPT) method, which fine-tunes pre-trained models with a modest number of parameters.<n>We convert raw point clouds into point embeddings by aggregating local geometry to capture spatial features followed by linear layers.<n>To calibrate self-attention across source domains of any modality to 3D, we introduce a prompt generator that shares weights with the point embedding module.
  arXiv  Detail & Related papers  (2025-08-30T06:02:21Z)
• A Moment Matching-Based Method for Sparse and Noisy Point Cloud Registration [8.121132773789652]
  We propose a registration framework based on moment matching.<n>Experiments on synthetic and real-world datasets show that our approach achieves higher accuracy and robustness than existing methods.<n>The proposed method significantly improves the localization performance and achieves results comparable to LiDAR-based systems.
  arXiv  Detail & Related papers  (2025-08-04T08:31:53Z)
• Rectified Point Flow: Generic Point Cloud Pose Estimation [33.190452313116936]
  We introduce Rectified Point Flow, a unified parameterization that formulates pairwise point cloud registration and multi-part shape assembly as a single conditional generative problem.<n>Our method learns a continuous point-wise velocity field that transports noisy points toward their target positions, from which part poses are recovered.
  arXiv  Detail & Related papers  (2025-06-05T17:36:03Z)
• DiffCom: Decoupled Sparse Priors Guided Diffusion Compression for Point Clouds [54.96190721255167]
  Lossy compression relies on an autoencoder to transform a point cloud into latent points for storage.<n>We propose a diffusion-based framework guided by sparse priors that achieves high reconstruction quality, especially at lows.
  arXiv  Detail & Related papers  (2024-11-21T05:41:35Z)
• A Consistency-Aware Spot-Guided Transformer for Versatile and Hierarchical Point Cloud Registration [9.609585217048664]
  We develop a consistency-aware spot-guided Transformer (CAST) CAST incorporates a spot-guided cross-attention module to avoid interfering with irrelevant areas. A lightweight fine matching module for both sparse keypoints and dense features can estimate the transformation accurately.
  arXiv  Detail & Related papers  (2024-10-14T08:48:25Z)
• Mask4Former: Mask Transformer for 4D Panoptic Segmentation [13.99703660936949]
  Mask4Former is the first transformer-based approach unifying semantic instance segmentation and tracking. Our model directly predicts semantic instances their temporal associations without relying on hand-crafted non-learned association strategies. Mask4Former achieves a new state-of-the-art on the SemanticTITI test set with a score of 68.4 LSTQ.
  arXiv  Detail & Related papers  (2023-09-28T03:30:50Z)
• Deep Confidence Guided Distance for 3D Partial Shape Registration [14.315501760755609]
  We present a novel non-iterative learnable method for partial-to-partial 3D shape registration. We present Confidence Guided Distance Network (CGD-net), where we fuse learnable similarity between point embeddings and spatial distance between point clouds.
  arXiv  Detail & Related papers  (2022-01-27T08:40:05Z)
• SPU-Net: Self-Supervised Point Cloud Upsampling by Coarse-to-Fine Reconstruction with Self-Projection Optimization [52.20602782690776]
  It is expensive and tedious to obtain large scale paired sparse-canned point sets for training from real scanned sparse data. We propose a self-supervised point cloud upsampling network, named SPU-Net, to capture the inherent upsampling patterns of points lying on the underlying object surface. We conduct various experiments on both synthetic and real-scanned datasets, and the results demonstrate that we achieve comparable performance to the state-of-the-art supervised methods.
  arXiv  Detail & Related papers  (2020-12-08T14:14:09Z)
• DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution [136.7261709896713]
  We propose a data-driven approach that generates the appropriate convolution kernels to apply in response to the nature of the instances. The proposed method achieves promising results on both ScanetNetV2 and S3DIS. It also improves inference speed by more than 25% over the current state-of-the-art.
  arXiv  Detail & Related papers  (2020-11-26T14:56:57Z)
• Quaternion Equivariant Capsule Networks for 3D Point Clouds [58.566467950463306]
  We present a 3D capsule module for processing point clouds that is equivariant to 3D rotations and translations. We connect dynamic routing between capsules to the well-known Weiszfeld algorithm. Based on our operator, we build a capsule network that disentangles geometry from pose.
  arXiv  Detail & Related papers  (2019-12-27T13:51:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.