Abstract. Static Terrestrial Laser Scanning (TLS) and SLAM-based Portable Mobile Laser Scanning (PMLS) are increasingly adopted in Cultural Heritage (CH) documentation, but their suitability for Historic/Heritage Building Information Modelling (HBIM) depends on both data quality and acquisition conditions. This paper compares a 2022 TLS survey and a 2025 handheld PMLS survey of San Giacomo Church (Como, Italy) to assess whether the latter can reliably support HBIM-oriented documentation. The methodology combines dataset-level comparison and ROI-based analysis on four stable architectural elements: apsis, pillar, timber roof truss, and central dome. Three complementary metrics were used: a local density proxy, a scale-dependent coverage ratio, and M3C2 distance statistics for geometric agreement. Results show that PMLS is consistently less dense than TLS but remains effective for 1:100 scale documentation and, in several cases, also for 1:50. Statistics on M3C2 distance remain generally within centimetric ranges, indicating good local agreement where surfaces are effectively observed. The study demonstrates that sensor suitability not only depends on the geometric complexity but also on sensor-to-surface distance, visibility, and acquisition geometry, supporting hybrid TLS–PMLS workflows for CH HBIM.
M. Garramone, L. Barazzetti, M. Scaioni· The International Archives o...· 0 citations
Abstract. Semantic segmentation of remote sensing imagery (RSI) is essential for urban mapping, land-use monitoring, and many other domains. However, pixel-level annotation is expensive, making weakly supervised semantic segmentation (WSSS) that relies on image-level labels an attractive alternative. Pre-trained models provide strong priors from large-scale learned representations, making them beneficial for WSSS. However, when kept frozen, they often produce sparse and misaligned class activation maps (CAMs) due to domain gaps and static inference. We propose a lightweight and efficient framework that integrates CLIP and DINO foundation models to address three challenges: (i) semantic misalignment between generic text prompts and RSI-specific visuals; (ii) static CAM quality; and (iii) incomplete object coverage. Our design includes: (1) a Textual Prototype-Aware Enrichment (TPE) module that builds an RS-specific knowledge base using large language model (LLM)-generated descriptions to enrich text prompts; (2) a Unified Semantic Relation Mining (USR) module that fuses learnable adapter features with CLIP attention and DINO affinity for online CAM refinement; and (3) a Visual Prototype-Aware Enrichment (VPE) module, which maintains momentum visual prototypes to complete regions and sharpen boundaries. By freezing the CLIP and DINO backbones and optimizing only lightweight adapter and decoder modules, the proposed framework reduces the number of trainable parameters while achieving competitive performance. Experimental on iSAID and ISPRS Potsdam datasets demonstrate the effectiveness of the proposed framework, achieving 38.01% mIoU on iSAID dataset and 47.01% mIoU with 66.89% overall accuracy on Potsdam dataset.
Xin Li, Nicola Genzano, M. Gianinetto et al.· ISPRS Annals of the Photogra...· 0 citations