EZ-SP: Fast and Lightweight Superpoint-Based 3D Segmentation

• URL: http://arxiv.org/abs/2512.00385v1
• Date: Sat, 29 Nov 2025 08:21:51 GMT
• Title: EZ-SP: Fast and Lightweight Superpoint-Based 3D Segmentation
• Abstract summary: We propose a learnable, fully GPU partitioning algorithm that generates geometrically and semantically coherent superpoints 13$times$ faster than prior methods.<n>Our module is compact (under 60k parameters), trains in under 20 minutes with a differentiable surrogate loss, and requires no handcrafted features.<n>With 72$times$ faster inference and 120$times$ fewer parameters, EZ-SP matches the accuracy of point-based SOTA models across three domains.
• Score: 14.136068396056869
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Superpoint-based pipelines provide an efficient alternative to point- or voxel-based 3D semantic segmentation, but are often bottlenecked by their CPU-bound partition step. We propose a learnable, fully GPU partitioning algorithm that generates geometrically and semantically coherent superpoints 13$\times$ faster than prior methods. Our module is compact (under 60k parameters), trains in under 20 minutes with a differentiable surrogate loss, and requires no handcrafted features. Combine with a lightweight superpoint classifier, the full pipeline fits in $<$2 MB of VRAM, scales to multi-million-point scenes, and supports real-time inference. With 72$\times$ faster inference and 120$\times$ fewer parameters, EZ-SP matches the accuracy of point-based SOTA models across three domains: indoor scans (S3DIS), autonomous driving (KITTI-360), and aerial LiDAR (DALES). Code and pretrained models are accessible at github.com/drprojects/superpoint_transformer.

Related papers

• TorR: Towards Brain-Inspired Task-Oriented Reasoning via Cache-Oriented Algorithm-Architecture Co-design [8.718677293812712]
  Task-oriented object detection atop CLIP offers open-vocabulary, prompt-driven semantics, yet dense per-window computation and heavy memory traffic hinder real-time, power-limited edge deployment.<n>We present emphTorR, a brain-inspired textbfalgorithm--architecture co-design that textbfreplaces CLIP-style dense alignment with a hyperdimensional (HDC) associative reasoner and turns temporal coherence into reuse.
  arXiv  Detail & Related papers  (2026-03-24T06:50:32Z)
• HARP-NeXt: High-Speed and Accurate Range-Point Fusion Network for 3D LiDAR Semantic Segmentation [39.58684038370709]
  LiDAR semantic segmentation is crucial for autonomous vehicles and mobile robots.<n>Previous state-of-the-art methods often face a trade-off between accuracy and speed.<n>We introduce HARP-NeXt, a high-speed and accurate LiDAR semantic segmentation network.
  arXiv  Detail & Related papers  (2025-10-08T10:46:07Z)
• TSP3D: Text-guided Sparse Voxel Pruning for Efficient 3D Visual Grounding [74.033589504806]
  We propose an efficient multi-level convolution architecture for 3D visual grounding.<n>Our method achieves top inference speed and surpasses previous fastest method by 100% FPS.
  arXiv  Detail & Related papers  (2025-02-14T18:59:59Z)
• ViG: Linear-complexity Visual Sequence Learning with Gated Linear Attention [33.00435765051738]
  We introduce Gated Linear Attention (GLA) for vision, leveraging its superior hardware-awareness and efficiency. Our proposed model, ViG, offers a favorable trade-off in accuracy, parameters, and FLOPs on ImageNet and downstream tasks. ViG-T uses 5.2$times$ fewer FLOPs, saves 90% GPU memory, runs 4.8$times$ faster, and achieves 20.7% higher top-1 accuracy than DeiT-T.
  arXiv  Detail & Related papers  (2024-05-28T17:59:21Z)
• Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering [7.680851067579922]
  We introduce a highly efficient method for panoptic segmentation of large 3D point clouds. Our model is over $30$ times smaller than the best-competing method and trains up to $15$ times faster.
  arXiv  Detail & Related papers  (2024-01-12T17:10:52Z)
• PointOcc: Cylindrical Tri-Perspective View for Point-based 3D Semantic Occupancy Prediction [72.75478398447396]
  We propose a cylindrical tri-perspective view to represent point clouds effectively and comprehensively. Considering the distance distribution of LiDAR point clouds, we construct the tri-perspective view in the cylindrical coordinate system. We employ spatial group pooling to maintain structural details during projection and adopt 2D backbones to efficiently process each TPV plane.
  arXiv  Detail & Related papers  (2023-08-31T17:57:17Z)
• Fully $1\times1$ Convolutional Network for Lightweight Image Super-Resolution [79.04007257606862]
  Deep models have significant process on single image super-resolution (SISR) tasks, in particular large models with large kernel ($3times3$ or more) $1times1$ convolutions bring substantial computational efficiency, but struggle with aggregating local spatial representations. We propose a simple yet effective fully $1times1$ convolutional network, named Shift-Conv-based Network (SCNet)
  arXiv  Detail & Related papers  (2023-07-30T06:24:03Z)
• GeoTransformer: Fast and Robust Point Cloud Registration with Geometric Transformer [63.85771838683657]
  We study the problem of extracting accurate correspondences for point cloud registration. Recent keypoint-free methods have shown great potential through bypassing the detection of repeatable keypoints. We propose Geometric Transformer, or GeoTransformer for short, to learn geometric feature for robust superpoint matching.
  arXiv  Detail & Related papers  (2023-07-25T02:36:04Z)
• Efficient 3D Semantic Segmentation with Superpoint Transformer [7.680851067579922]
  We introduce a novel superpoint-based transformer architecture for efficient semantic segmentation of large-scale 3D scenes. Our method incorporates a fast algorithm to partition point clouds into a hierarchical superpoint structure, which makes our preprocessing 7 times faster than existing approaches. Our model can be trained on a single GPU in 3 hours for a fold of the S3DIS dataset, which is 7x to 70x fewer GPU-hours than the best-performing methods.
  arXiv  Detail & Related papers  (2023-06-13T18:03:05Z)
• DSVT: Dynamic Sparse Voxel Transformer with Rotated Sets [95.84755169585492]
  We present Dynamic Sparse Voxel Transformer (DSVT), a single-stride window-based voxel Transformer backbone for outdoor 3D perception. Our model achieves state-of-the-art performance with a broad range of 3D perception tasks.
  arXiv  Detail & Related papers  (2023-01-15T09:31:58Z)
• Superpoint Transformer for 3D Scene Instance Segmentation [7.07321040534471]
  This paper proposes a novel end-to-end 3D instance segmentation method based on Superpoint Transformer, named as SPFormer. It groups potential features from point clouds into superpoints, and directly predicts instances through query vectors. It exceeds compared state-of-the-art methods by 4.3% on ScanNetv2 hidden test set in terms of mAP and keeps fast inference speed (247ms per frame) simultaneously.
  arXiv  Detail & Related papers  (2022-11-28T20:52:53Z)
• FIDNet: LiDAR Point Cloud Semantic Segmentation with Fully Interpolation Decoding [5.599306291149907]
  Projecting the point cloud on the 2D spherical range image transforms the LiDAR semantic segmentation to a 2D segmentation task on the range image. We propose a new projection-based LiDAR semantic segmentation pipeline that consists of a novel network structure and an efficient post-processing step. Our pipeline achieves the best performance among all projection-based methods with $64 times 2048$ resolution and all point-wise solutions.
  arXiv  Detail & Related papers  (2021-09-08T17:20:09Z)
• Dynamic Convolution for 3D Point Cloud Instance Segmentation [146.7971476424351]
  We propose an approach to instance segmentation from 3D point clouds based on dynamic convolution. We gather homogeneous points that have identical semantic categories and close votes for the geometric centroids. The proposed approach is proposal-free, and instead exploits a convolution process that adapts to the spatial and semantic characteristics of each instance.
  arXiv  Detail & Related papers  (2021-07-18T09:05:16Z)
• Learning Semantic Segmentation of Large-Scale Point Clouds with Random Sampling [52.464516118826765]
  We introduce RandLA-Net, an efficient and lightweight neural architecture to infer per-point semantics for large-scale point clouds. The key to our approach is to use random point sampling instead of more complex point selection approaches. Our RandLA-Net can process 1 million points in a single pass up to 200x faster than existing approaches.
  arXiv  Detail & Related papers  (2021-07-06T05:08:34Z)

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.