Advancements in image processing and deep learning offer considerable opportunities for automated defect assessment in civil structures. However, these systems cannot work interactively with human inspectors. Mixed reality (MR) can be adopted to address this by involving inspectors in various stages of the assessment process. This paper integrates You Only Look Once (YOLO) v5n and YOLO v5m with the Canny algorithm for real-time concrete crack detection and skeleton extraction with a Microsoft HoloLens 2 MR device. The YOLO v5n demonstrates a superior mean average precision (mAP) 0.5 and speed, while YOLO v5m achieves the highest mAP 0.5 0.95 among the other YOLO v5 structures. The Canny algorithm also outperforms the Sobel and Prewitt edge detectors with the highest F1 score. The developed MR-based system could not only be employed for real-time defect assessment but also be utilized for the automatic recording of the location and other specifications of the cracks for further analysis and future re-inspections.

Pole-like objects (PLOs) are important street assets in urban environments, yet current deep learning methods often underperform in their segmentation compared to other objects. The main challenge is determining the right kernel size to effectively understand the unique structure of PLOs with an appropriate receptive field. In this study, we improve the segmentation performance of PLOs by optimizing the kernel size in a KPConv-based network. Our experiments show that kernel size of 9 yields an Intersection over Union (IoU) of 95.02% on the Parkville-3D dataset. We also develop a post-processing approach that transforms semantic segmentation outputs into panoptic segmentation results, enabling accurate detection of individual PLO instances. Furthermore, qualitative tests on an independent, unlabelled point cloud dataset from a different urban area demonstrate that our method consistently achieves accurate segmentation.

Pole-like objects (PLOs) are critical for urban infrastructure mapping but remain challenging to detect in mobile LiDAR scans without large, annotated datasets. We introduce SynPole, the first synthetic zero-shot framework for PLO classification that requires only a single CAD model of the target pole. SynPole consists of a lightweight, object-centric LiDAR simulator to generate realistic synthetic scans from a 3D model, and a training-free, non-parametric encoder that extracts fixed-length features from support samples. At inference, object proposals can obtained via any unsupervised, class-agnostic segmentation and classified by the similarity to support features. Experimental evaluation shows that SynPole achieves a 71.4% Intersection-over-Union (IoU) on PLO, demonstrating that our method matches the performance of state-of-the-art, fully supervised baselines while reducing huge annotation effort.