SOURCE-LINKED INTELLIGENCE

Partition-Invariant Tuning for 3D Scene Understanding

arXiv · AI, language, vision and robotics · article · Sep 11, 2026 · UTC

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

Read original source ↗ Open in workspace

recordType

paper

region

Global