PulseAugur
EN
LIVE 22:08:57

New PCFootprint dataset advances building footprint extraction from LiDAR

Researchers have introduced PCFootprint, a new large-scale dataset designed for extracting vectorized building footprints from aerial LiDAR point clouds. This dataset, comprising 33,000 tiles derived from the Estonian Land and Spatial Development Board, aims to overcome limitations of optical imagery, such as occlusions and lack of elevation data. PCFootprint includes a cross-domain test set to evaluate generalization and establishes benchmarks for evaluating existing methods, highlighting challenges like data imbalance and noise. AI

IMPACT This dataset could improve building modeling and urban scene understanding by enabling more robust footprint extraction from LiDAR data.

RANK_REASON The cluster describes the release of a new academic dataset and benchmark for a computer vision task.

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

New PCFootprint dataset advances building footprint extraction from LiDAR

COVERAGE [2]

  1. arXiv cs.CV TIER_1 English(EN) · Haoyuan Shen, Kuihao Wang, Ruisheng Wang, Yujun Liu ·

    PCFootprint: A Large-Scale Dataset and Benchmark for Vectorized Building Footprint Extraction from Aerial LiDAR Point Clouds

    arXiv:2606.20455v1 Announce Type: new Abstract: Building footprint extraction is a fundamental task in photogrammetry, remote sensing, and computer vision. Recent image-based methods have achieved remarkable progress in extracting vectorized footprints from high-resolution optica…

  2. arXiv cs.CV TIER_1 English(EN) · Yujun Liu ·

    PCFootprint: A Large-Scale Dataset and Benchmark for Vectorized Building Footprint Extraction from Aerial LiDAR Point Clouds

    Building footprint extraction is a fundamental task in photogrammetry, remote sensing, and computer vision. Recent image-based methods have achieved remarkable progress in extracting vectorized footprints from high-resolution optical imagery. However, optical imagery inherently s…