PointTPA: Dynamic Network Parameter Adaptation for 3D Scene Understanding

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Siyuan, Zheng, Chaoqun, Zhou, Xin, Feng, Tianrui, Liang, Dingkang, Bai, Xiang
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917386793254912
author Liu, Siyuan
Zheng, Chaoqun
Zhou, Xin
Feng, Tianrui
Liang, Dingkang
Bai, Xiang
author_facet Liu, Siyuan
Zheng, Chaoqun
Zhou, Xin
Feng, Tianrui
Liang, Dingkang
Bai, Xiang
contents Scene-level point cloud understanding remains challenging due to diverse geometries, imbalanced category distributions, and highly varied spatial layouts. Existing methods improve object-level performance but rely on static network parameters during inference, limiting their adaptability to dynamic scene data. We propose PointTPA, a Test-time Parameter Adaptation framework that generates input-aware network parameters for scene-level point clouds. PointTPA adopts a Serialization-based Neighborhood Grouping (SNG) to form locally coherent patches and a Dynamic Parameter Projector (DPP) to produce patch-wise adaptive weights, enabling the backbone to adjust its behavior according to scene-specific variations while maintaining a low parameter overhead. Integrated into the PTv3 structure, PointTPA demonstrates strong parameter efficiency by introducing two lightweight modules of less than 2% of the backbone's parameters. Despite this minimal parameter overhead, PointTPA achieves 78.4% mIoU on ScanNet validation, surpassing existing parameter-efficient fine-tuning (PEFT) methods across multiple benchmarks, highlighting the efficacy of our test-time dynamic network parameter adaptation mechanism in enhancing 3D scene understanding. The code is available at https://github.com/H-EmbodVis/PointTPA.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04933
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PointTPA: Dynamic Network Parameter Adaptation for 3D Scene Understanding
Liu, Siyuan
Zheng, Chaoqun
Zhou, Xin
Feng, Tianrui
Liang, Dingkang
Bai, Xiang
Computer Vision and Pattern Recognition
Scene-level point cloud understanding remains challenging due to diverse geometries, imbalanced category distributions, and highly varied spatial layouts. Existing methods improve object-level performance but rely on static network parameters during inference, limiting their adaptability to dynamic scene data. We propose PointTPA, a Test-time Parameter Adaptation framework that generates input-aware network parameters for scene-level point clouds. PointTPA adopts a Serialization-based Neighborhood Grouping (SNG) to form locally coherent patches and a Dynamic Parameter Projector (DPP) to produce patch-wise adaptive weights, enabling the backbone to adjust its behavior according to scene-specific variations while maintaining a low parameter overhead. Integrated into the PTv3 structure, PointTPA demonstrates strong parameter efficiency by introducing two lightweight modules of less than 2% of the backbone's parameters. Despite this minimal parameter overhead, PointTPA achieves 78.4% mIoU on ScanNet validation, surpassing existing parameter-efficient fine-tuning (PEFT) methods across multiple benchmarks, highlighting the efficacy of our test-time dynamic network parameter adaptation mechanism in enhancing 3D scene understanding. The code is available at https://github.com/H-EmbodVis/PointTPA.
title PointTPA: Dynamic Network Parameter Adaptation for 3D Scene Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.04933