PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 17, 2026AI0 citationsOpen Access

SPICD-Net: A Siamese PointNet Framework for Autonomous Indoor Change Detection in 3D LiDAR Point Clouds

View Full Paper
DŠDalibor ŠeljmešiVBVladimir BrtkaVIVelibor Ilić

Key Points

  • The aim is to develop a reliable system for detecting indoor changes using 3D LiDAR data without manual annotation.
  • Developed a Siamese PointNet framework for classifying tile pairs into no-change, changed, and inconsistent categories.
  • Implemented a synthetic anomaly injection strategy to align training with real-time processing needs.
  • Introduced a stochastic-gated Chamfer-statistics branch to enhance geometric feature analysis under hardware constraints.
  • Achieved Precision = 0.86, Recall = 0.82, F1-score = 0.84, and Accuracy = 0.96 on 14 simulation experiments.
  • No false positives were recorded in the no-change baseline, and mean inference time was 22.4 seconds per 172-tile map.
  • Limited real-world tests resulted in Precision = 0.583, Recall = 1.000, and F1 = 0.737 for an unseen room.

Abstract

Reliable change detection in indoor environments remains a challenge for autonomous robotic systems using 3D LiDAR. Existing methods often require manual annotation, computationally intensive architectures, or focus on outdoor scenes. This paper presents SPICD-Net, a lightweight Siamese PointNet framework for indoor 3D change detection trained exclusively on synthetically generated anomalies, eliminating manual labeling. The framework offers three deployment-oriented contributions: a three-class Siamese formulation separating no-change, changed, and geometrically inconsistent tile pairs; a pre-FPS anomaly injection strategy that aligns synthetic training with inference-time preprocessing; and a stochastic-gated Chamfer-statistics branch that complements learned embeddings with explicit geometric cues under consumer-grade hardware constraints. Evaluated on 14 controlled simulation experiments in an indoor corridor dataset, SPICD-Net achieved aggregated Precision = 0.86, Recall = 0.82, F1-score = 0.84, and Accuracy = 0.96, with zero false positives in the no-change baseline and mean inference time of 22.4 s for a 172-tile map on a single consumer GPU. Additional robustness experiments identified registration accuracy as the main operational prerequisite. A limited real-world validation in one unseen room (four scans, 67 tiles) achieved Precision = 0.583, Recall = 1.000, and F1 = 0.737.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Šeljmeši et al. (2026) studied this question.

synapsesocial.com/papers/69e1d0165cdc762e9d8592e8https://doi.org/10.3390/ai7040141
Ask AI
Helpful
Bookmark
Share
View Full Paper