Saved in:
Bibliographic Details
Main Authors: Bekar, Batuhan Arda, Sarı, Can, Gülkan, Hüseyin Can, Özcan, Barış
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.15088
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913129879830528
author Bekar, Batuhan Arda
Sarı, Can
Gülkan, Hüseyin Can
Özcan, Barış
author_facet Bekar, Batuhan Arda
Sarı, Can
Gülkan, Hüseyin Can
Özcan, Barış
contents We present SAGE3D, a hybrid Transformer-based model for corner detection in airborne LiDAR point clouds. We propose a multi-stage solution built on a hierarchical encoder-decoder architecture that progressively downsamples point clouds through Set Abstraction layers and recovers per-point predictions via Feature Propagation. We introduce two innovations: Soft-Guided Attention, which injects ground-truth corner labels as a log-prior into attention logits during training to improve precision; then an Excitatory Graph Neural Network positioned at strategic resolutions in the hierarchy, employing positive-only message passing where high-confidence corners reinforce predictions through learned boosting, optimizing for recall. The hierarchical design enables multi-scale feature extraction while our guided attention and excitatory modules ensure corner signals are amplified rather than diluted across scales.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15088
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SAGE3D: Soft-guided attention and graph excitation for 3D point cloud corner detection
Bekar, Batuhan Arda
Sarı, Can
Gülkan, Hüseyin Can
Özcan, Barış
Computer Vision and Pattern Recognition
We present SAGE3D, a hybrid Transformer-based model for corner detection in airborne LiDAR point clouds. We propose a multi-stage solution built on a hierarchical encoder-decoder architecture that progressively downsamples point clouds through Set Abstraction layers and recovers per-point predictions via Feature Propagation. We introduce two innovations: Soft-Guided Attention, which injects ground-truth corner labels as a log-prior into attention logits during training to improve precision; then an Excitatory Graph Neural Network positioned at strategic resolutions in the hierarchy, employing positive-only message passing where high-confidence corners reinforce predictions through learned boosting, optimizing for recall. The hierarchical design enables multi-scale feature extraction while our guided attention and excitatory modules ensure corner signals are amplified rather than diluted across scales.
title SAGE3D: Soft-guided attention and graph excitation for 3D point cloud corner detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.15088