Partition-Invariant Tuning for 3D Scene Understanding

• URL: http://arxiv.org/abs/2609.12473v1
• Date: Fri, 11 Sep 2026 06:14:48 GMT
• Title: Partition-Invariant Tuning for 3D Scene Understanding
• Abstract summary: We propose PointPiT, a partition-invariant tuning framework for scene-level point clouds.<n>PointPiT integrates local geometric patterns with global scene context to suppress partition-induced representation shifts.<n>Experiments demonstrate that PointPiT achieves competitive or even superior performance to full fine-tuning with less than 1% of backbone's parameters.
• Score: 10.734083086241457
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Scene-level point cloud understanding remains challenging due to diverse geometries and spatial layouts. While pre-trained 3D point cloud foundation models (PFMs) offer strong transferability, full fine-tuning (FFT) incurs substantial computational and storage costs. Parameter-efficient fine-tuning (PEFT) provides a promising alternative, but existing PEFT methods largely focus on object-level point clouds and overlook serialization-induced partition variations in large-scale scenes. To address this issue, we propose PointPiT, a partition-invariant tuning framework for scene-level point clouds. Specifically, a Scene-aware Structural Adapter (SSA) integrates local geometric patterns with global scene context to mitigate partition-induced representation shifts. Moreover, Gradient Subspace Optimization (GSO) selects informative and partition-stable update directions, suppressing partition-dependent variations during optimization. Extensive experiments across multiple scene-level benchmarks demonstrate that PointPiT achieves competitive or even superior performance to full fine-tuning with less than 1% of backbone's parameters, while achieving consistent state-of-the-art performance among representative PEFT methods.

Related papers

• Point Ladder Tuning: Parameter-Efficient Hierarchical Adaptation for 3D Point Cloud Understanding [4.750456629699177]
  Point Ladder Tuning (PLT) is a PEFT framework that performs hierarchical, instance-conditioned adaptation while keeping the backbone frozen.<n>For dense prediction, we introduce a lightweight segmentation head that progressively upsamples fused features.<n>Experiments on classification and dense prediction show that consistently surpasses prior PEFT baselines with minimal parameters.
  arXiv  Detail & Related papers  (2026-07-21T15:06:37Z)
• PointTPA: Dynamic Network Parameter Adaptation for 3D Scene Understanding [51.98954431091292]
  We propose a Test-time Adaptation framework that generates input-aware network parameters for scene-level point clouds.<n>PointTPA achieves 78.4% mIoU on ScanNet validation, surpassing existing parameter-efficient fine-tuning methods.
  arXiv  Detail & Related papers  (2026-04-06T17:59:58Z)
• 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)
• On Geometry-Enhanced Parameter-Efficient Fine-Tuning for 3D Scene Segmentation [52.96632954620623]
  We introduce a novel geometry-aware PEFT module specifically designed for 3D point cloud transformers.<n>Our approach sets a new benchmark for efficient, scalable, and geometry-aware fine-tuning of large-scale 3D point cloud models.
  arXiv  Detail & Related papers  (2025-05-28T15:08:36Z)
• PIVOT-Net: Heterogeneous Point-Voxel-Tree-based Framework for Point Cloud Compression [8.778300313732027]
  We propose a heterogeneous point cloud compression (PCC) framework. We unify typical point cloud representations -- point-based, voxel-based, and tree-based representations -- and their associated backbones. We augment the framework with a proposed context-aware upsampling for decoding and an enhanced voxel transformer for feature aggregation.
  arXiv  Detail & Related papers  (2024-02-11T16:57:08Z)
• 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)
• 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)
• 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)

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.