Aerial Triangulation With Learning-Based Tie Points

Abstract

Abstract. Aerial triangulation (AT) has reached outstanding progress in the last decades, and now fully automated solutions for nadir and oblique images are available. Usually, image correspondences (tie points) are found using hand-crafted methods, such as SIFT or its variants. But in the last years, there were many investigations and developments to promote the use of machine and deep learning solutions within the photogrammetric processing pipeline. The paper explores learning-based methods for the extraction of tie points in aerial image blocks. Image correspondences are used to perform aerial triangulation (AT) and successively generate dense point clouds. Two different datasets are used to compare conventional hand-crafted detector/descriptor methods with respect to learning-based methods. Accuracy analyses are performed using GCPs as well as ground truth LiDAR point clouds. Results confirm the potential of learningbased methods in finding reliable image correspondences in the aerial block, still showing space for improvements due to camera rotations.

Evaluating Learning-based Tie Point Matching for Geometric Processing of Off-Track Satellite Stereo

“…In the context of classical photogrammetric datasets, characterized by single-sensor acquisitions within a limited timeframe and substantial image overlap, the adoption of learned approaches offers minimal advantages and, in certain instances, may even result in reduced accuracy, as highlighted by Remondino et al (2021). The advantage is instead evident in challenging multi-temporal datasets (Maiwald et al, 2021;Morelli et al, 2022) or under different viewing angles (Ioli et al, 2023). It is noteworthy that these approaches have inherent constraints, including the ability to execute predictions solely on images of limited dimensions determined by GPU capabilities, as well as limitations in rotation and scale invariance, as observed in Marelli et al (2023).…”

“…LightGlue has been chosen, since it is an optimized version of SuperGlue with a more permissive license. These algorithms are available in the DIM (Deep-Image-Matching) library (Morelli et al, , 2022, which has been designed to process large-format images as the chosen 40 satellite image pairs detailed in Section 4.1.…”

“…Deep learning (DL) based approaches are showing consistently progresses in image-matching problems and benchmarks (Jin et al, 2021;Remondino et al, 2022). Owning to its ability of learning complex features by samples, it was shown to be effective in addressing correspondence problem between images with drastic differences in scale, illumination, and colorimetry (Morelli et al, 2022;Morelli et al, 2024). However, their ability in addressing the compounded challenges in satellite off-track stereo pairs remain untested.…”

Deep-Image-Matching: A Toolbox for Multiview Image Matching of Complex Scenarios

“…Over the last decade, there has been a proliferation of deep learning (DL) approaches for feature extraction and matching (Chen et al, 2021;Jin et al 2021;Yao et al, 2021) that aim to overcome these limitations and they have demonstrated resilience against varying illumination conditions, multitemporal datasets, wide baselines, and significantly different view angles. Recently, several works have proved the effectiveness of DL approaches in challenging scenarios, including glacier monitoring with wide camera baselines (Ioli et al, 2023a, Ioli et al, 2023b, multi-temporal image matching (Maiwald et al, 2023), multi-temporal co-registration problems (Maiwald et al, 2021;, VO and SLAM (Morelli et al, 2023), aerial triangulation (Remondino et al, 2022) and in terrestrial laser scanning point cloud registration (Markiewicz et al, 2023). However, well known limitations of DL approaches are their computational complexity, limited scale and rotation invariance of the descriptors and their application on high-resolution images.…”

Digitizing Historical Aerial Images: Evaluation of the Effects of Scanning Quality on Aerial Triangulation and Dense Image Matching

“…Although these methods proved to be effective for image matching tasks, in the last years, deep-learning approaches and hybrid processing pipelines have also emerged and been demonstrated to often overcome common limitations of traditional methods, e.g., multi-temporal data, radiometric changes, etc. [30][31][32][33][34]. Learning-based methods started to be applied to historical images for 4D urban reconstruction purposes [26,35,36].…”