Large-Scale DTM Generation From Satellite Data

Duan, Liuyun; Desbrun, Mathieu; Giraud, Anne; Trastour, Frederic; Laurore, L.

doi:10.1109/cvprw.2019.00185

Cited by 9 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…and artificial (buildings, construction, etc.). The typical process composes of 2 steps, ground filtering and surface interpolation [8], [10]. Ground filtering, or terrain extraction, identifies the ground portions from among other entities (water, trees, buildings, etc.)…”

Section: A Dtm Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Learning Digital Terrain Models From Point Clouds: ALS2DTM Dataset and Rasterization-Based GAN

Guiotte

Pham

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Despite the popularity of deep neural networks in various domains, the extraction of digital terrain models (DTMs) from airborne laser scanning (ALS) point clouds is still challenging. This might be due to the lack of dedicated large-scale annotated dataset and the data-structure discrepancy between point clouds and DTMs. To promote data-driven DTM extraction, this paper collects from open sources a large-scale dataset of ALS point clouds and corresponding DTMs with various urban, forested, and mountainous scenes. A baseline method is proposed as the first attempt to train a Deep neural network to extract digital Terrain models directly from ALS point clouds via Rasterization techniques, coined DeepTerRa. Extensive studies with well-established methods are performed to benchmark the dataset and analyze the challenges in learning to extract DTM from point clouds. The experimental results show the interest of the agnostic data-driven approach, with sub-metric error level compared to methods designed for DTM extraction. The data and source code is provided at https://lhoangan.github.io/deepterra/ for reproducibility and further similar research.

show abstract

Section: A Dtm Extractionmentioning

confidence: 99%

“…DTM extraction can be achieved from satellite or aerial images [6], [7], DSMs [8], or point clouds [9], [10]. Among the possibilities, airborne laser scanning (ALS) point clouds, with rich geometry information (c.f .…”

Section: Introductionmentioning

confidence: 99%

Learning Digital Terrain Models From Point Clouds: ALS2DTM Dataset and Rasterization-Based GAN

Guiotte

Pham

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The class label matrix includes pixel labels similar to the outputs in the semantic segmentation task [8], [9]. A set of morphological operations [52] (e.g., denoising, erosion, opening, and closing) are then applied on to smooth and remove noise from these matrices. The refined matrices are therefore obtained from 5 class label matrices which correspond to 5 scale levels.…”

Section: Oriented Box Determinationmentioning

confidence: 99%

Oriented Bounding Boxes for Small and Freely Rotated Objects

Zand,

Etemad,

Greenspan

2021

Preprint

View full text Add to dashboard Cite

A novel object detection method is presented that handles freely rotated objects of arbitrary sizes, including tiny objects as small as 2 × 2 pixels. Such tiny objects appear frequently in remotely sensed images, and present a challenge to recent object detection algorithms. More importantly, current object detection methods have been designed originally to accommodate axis-aligned bounding box detection, and therefore fail to accurately localize oriented boxes that best describe freely rotated objects. In contrast, the proposed CNN-based approach uses potential pixel information at multiple scale levels without the need for any external resources, such as anchor boxes. The method encodes the precise location and orientation of features of the target objects at grid cell locations. Unlike existing methods which regress the bounding box location and dimension, the proposed method learns all the required information by classification, which has the added benefit of enabling oriented bounding box detection without any extra computation. It thus infers the bounding boxes only at inference time by finding the minimum surrounding box for every set of the same predicted class labels. Moreover, a rotation-invariant feature representation is applied to each scale, which imposes a regularization constraint to enforce covering the 360 degree range of in-plane rotation of the training samples to share similar features. Evaluations on the xView and DOTA datasets show that the proposed method uniformly improves performance over existing state-of-the-art methods.

show abstract

“…Both rely on knowledge of ground level in the DTM. For the lidar data used in our experiments, DTM layers were produced by professional surveyors with manual edits, but automated methods for ground classification in lidar and even in DSMs produced using satellite images also work well [8].…”

Section: Supervisionmentioning

confidence: 99%

Learning Geocentric Object Pose in Oblique Monocular Images

Christie¹,

Abujder²,

Foster³

et al. 2020

Preprint

View full text Add to dashboard Cite

An object's geocentric pose, defined as the height above ground and orientation with respect to gravity, is a powerful representation of real-world structure for object detection, segmentation, and localization tasks using RGBD images. For close-range vision tasks, height and orientation have been derived directly from stereo-computed depth and more recently from monocular depth predicted by deep networks. For long-range vision tasks such as Earth observation, depth cannot be reliably estimated with monocular images. Inspired by recent work in monocular height above ground prediction and optical flow prediction from static images, we develop an encoding of geocentric pose to address this challenge and train a deep network to compute the representation densely, supervised by publicly available airborne lidar. We exploit these attributes to rectify oblique images and remove observed object parallax to dramatically improve the accuracy of localization and to enable accurate alignment of multiple images taken from very different oblique viewpoints. We demonstrate the value of our approach by extending two large-scale public datasets for semantic segmentation in oblique satellite images. All of our data and code are publicly available 1 .

show abstract

Large-Scale DTM Generation From Satellite Data

Cited by 9 publications

References 16 publications

Learning Digital Terrain Models From Point Clouds: ALS2DTM Dataset and Rasterization-Based GAN

Learning Digital Terrain Models From Point Clouds: ALS2DTM Dataset and Rasterization-Based GAN

Oriented Bounding Boxes for Small and Freely Rotated Objects

Learning Geocentric Object Pose in Oblique Monocular Images

Contact Info

Product

Resources

About