iVISION-2DCD: A Long-Term Change Detection Dataset for Large-Scale Outdoor Construction Monitoring
Dayou Mao, Yuchen Lin, Ashkan Ebadi, John S. Zelek, Alexander Wong, Yuhao Chen
AI summary
Problem
Existing datasets for construction change detection lack long-term temporal coverage, diverse camera viewpoints, and outdoor environmental complexity, while real-world UAV data collection is constrained by safety limits and misaligned bi-temporal captures.
Approach
The authors synthesize a large-scale dataset from dense LiDAR point clouds using novel view generation techniques to create diverse, bi-temporally aligned camera perspectives, paired with semi-automated semantic segmentation to produce accurate change labels.
Key results
- Large-scale synthetic dataset spanning 32.3K m² and over 50 acquisition sessions across one year
- Novel view synthesis pipeline generating diverse viewpoints to overcome UAV safety constraints
- Semi-automated change label generation from 3D semantic point clouds ensuring symmetric alignment
- Benchmark evaluation revealing state-of-the-art 2DCD algorithms struggle with oblique views and seasonal variations
Why it matters
It provides the computer vision and robotics communities with a critical benchmark for developing robust, viewpoint-agnostic change detection systems for large-scale outdoor construction automation.
Abstract
Automation in construction is essential for re- ducing costs and human errors in large-scale projects. We approach the construction progress monitoring from the aspect of detecting changes in construction sites. As construction buildings continue to evolve in geometry and appearance over time, change detection need to be performed from ar- bitrary camera viewpoints. This necessitates developing 2D Change Detection (2DCD) algorithms that operate robustly across diverse camera perspectives at construction sites. While developing and evaluating such systems is data-intensive, no open-source benchmark dataset exists at the intersection of 2D change detection and construction automation research. Data collection using Unmanned Aerial Vehicles (UAVs) is gaining its popularity in outdoor large-scale surveying. However, in active construction sites conducting drone missions equipped with high-end sensors imposes safety concerns. Flight trajectory and collected camera viewpoints can be significantly limited. To address this critical gap, we introduce iVISION-2DCD, a large- scale synthetically generated dataset from dense LiDAR point clouds with photorealistic input images and accurate ground truth annotations. Our dataset formally defines the problem of viewpoint-robust 2DCD at construction sites and captures the inherent complexities of real-world deployment. In this paper, we present our systematic methodology for synthetic data generation, developing novel view synthesis techniques to over- come bi-temporal alignment and viewpoint diversity challenges, and implementing semi-automated semantic segmentation with change label generation while preserving challenging real-world cases. Benchmark evaluations using state-of-the-art 2DCD algo- rithms demonstrate that iVISION-2DCD poses novel research challenges for the computer vision and robotics communities.