Papers › SAT: Size-Aware Transformer for 3D Point Cloud Semantic Segmentation

SAT: Size-Aware Transformer for 3D Point Cloud Semantic Segmentation

17 Jan 2023arXiv:2301.06869archive 2025-07-28

Junjie Zhou, Yongping Xiong, Chinwai Chiu, Fangyu Liu, Xiangyang Gong

Transformer models have achieved promising performances in point cloud segmentation. However, most existing attention schemes provide the same feature learning paradigm for all points equally and overlook the enormous difference in size among scene objects. In this paper, we propose the Size-Aware Transformer (SAT) that can tailor effective receptive fields for objects of different sizes. Our SAT achieves size-aware learning via two steps: introduce multi-scale features to each attention layer and allow each point to choose its attentive fields adaptively. It contains two key designs: the Multi-Granularity Attention (MGA) scheme and the Re-Attention module. The MGA addresses two challenges: efficiently aggregating tokens from distant areas and preserving multi-scale features within one attention layer. Specifically, point-voxel cross attention is proposed to address the first challenge, and the shunted strategy based on the standard multi-head self attention is applied to solve the second. The Re-Attention module dynamically adjusts the attention scores to the fine- and coarse-grained features output by MGA for each point. Extensive experimental results demonstrate that SAT achieves state-of-the-art performances on S3DIS and ScanNetV2 datasets. Our SAT also achieves the most balanced performance on categories among all referred methods, which illustrates the superiority of modelling categories of different sizes. Our code and model will be released after the acceptance of this paper.

Paper PDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Point Cloud Segmentation Semantic Segmentation

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Semantic Segmentation	S3DIS Area5	SAT (FAT)	Number of params	N/A	#17 of 61	Archive leaderboard	report
Semantic Segmentation	S3DIS Area5	SAT (FAT)	mAcc	78.8	#17 of 61	Archive leaderboard	report
Semantic Segmentation	S3DIS Area5	SAT (FAT)	mIoU	72.6	#17 of 61	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position Encodings Adam Attention BPE Dense Connections Dropout Label Smoothing Layer Normalization Linear Layer Multi-Head Attention Position-Wise Feed-Forward Layer Re-Attention Module Residual Connection Softmax Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections