Three-dimensional (3D) reconstruction of structures and landscapes: A new point-and-line fusion method

Zhou, Ying; Wang, Lingling; Love, Peter; Ding, Lieyun; Zhou, Cheng

doi:10.1016/j.aei.2019.100961

Cited by 19 publications

(13 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In practice, Levenberg–Marquardt algorithm is usually used to iteratively optimize the minimum reprojection error [ 10 ], which is the most widely used nonlinear least square algorithm, and its iterative formula is as follows:

where λ is the weight parameter; when yλ is large, the above formula is the gradient descent method with a small step size, and when λ is small, it is the Gauss-Newton method.…”

Section: Methodsmentioning

confidence: 99%

“…Because the RGB⁃D image sequence is not an accurate result in the layout prediction of new layout instances, there are certain layout conflicts, such as the phenomenon that the functional areas across the wall and the functional areas overlap [ 10 ]. Therefore, in the layout planning of 3D indoor scenes, the occlusion edges in RGB⁃D images are calculated, mainly including no wall penetration, no overlap, and no door blocking.…”

Section: Methodsmentioning

confidence: 99%

“…TDR of the three-dimensional indoor scene based on RGB-d sensor spans many fields such as computer vision and robotics. The author of [ 10 ] analyzes the calibration of the Kinect v1 sensor and analyzes its depth image information in terms of accuracy and resolution. This paper studies how to use Kinect v1 to build a dense 3D map of the indoor environment for robot navigation, semantic mapping, and remote presentation.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

3D Indoor Scene Reconstruction and Layout Based on Virtual Reality Technology and Few-Shot Learning

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Indoor three-dimensional layout has a strong application background, such as virtual office three-dimensional layout planning, museum three-dimensional layout planning, and cave scene three-dimensional layout planning, which have been widely used in telecommuting, education, tourism, and other industries. In view of this, this paper proposes an indoor landscape reconstruction method based on VR (virtual reality) and draws indoor landscape information and images by using VR technology to generate an indoor landscape reconstruction panorama. A model is established to correct the distance error and reflectivity error of depth image, improve the accuracy of the depth image, and finally improve the accuracy of three-dimensional indoor scene TDR (three-dimensional reconstruction). In the process of optimizing layout, the Monte Carlo sampling method is used based on the Markov chain, and constraints are used as density functions to guide layout sampling and generate a number of reasonable scene layout suggestions in the iterative process of the sampler. Experiments show that this method can provide scientific and reasonable guidance to users’ scene layout and help them complete the furniture layout quickly.

show abstract

where λ is the weight parameter; when yλ is large, the above formula is the gradient descent method with a small step size, and when λ is small, it is the Gauss-Newton method.…”

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

3D Indoor Scene Reconstruction and Layout Based on Virtual Reality Technology and Few-Shot Learning

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…The disparity map output of the SGM, on the other hand, contains certain mismatches where the overlapping images are placed at different viewing angles. The generated 3D model using the patch‐based multi‐view stereo (PMVS) software (Furukawa and Ponce, 2009) or other approaches still has some flaws (Zhou et al., 2019). This is because some negative factors, such as spectral changes and blocked areas, cause more blunder errors on tall buildings (Noronha and Nevatia, 2001; Siddiqui et al., 2016; Mohammadi et al., 2019).…”

Section: Introductionmentioning

confidence: 99%

True Orthophoto Generation Based on Unmanned Aerial Vehicle Images Using Reconstructed Edge Points

Ebrahimikia

Ahmadabadian

2022

The Photogrammetric Record

View full text Add to dashboard Cite

After considering state‐of‐the‐art algorithms, this paper presents a novel method for generating true orthophotos from unmanned aerial vehicle (UAV) images of urban areas. The procedure consists of four steps: 2D edge detection in building regions, 3D edge graph generation, digital surface model (DSM) modification and, finally, true orthophoto and orthomosaic generation. The main contribution of this paper is concerned with the first two steps, in which deep‐learning approaches are used to identify the structural edges of the buildings and the estimated 3D edge points are added to the point cloud for DSM modification. Running the proposed method as well as four state‐of‐the‐art methods on two different datasets demonstrates that the proposed method outperforms the existing orthophoto improvement methods by up to 50% in the first dataset and by 70% in the second dataset by reducing true orthophoto distortion in the structured edges of the buildings.

show abstract

“…A LiDAR system actively emits laser pulses to a target and receives the reflected signals, and the point clouds are usually directly recorded by LiDAR devices and saved in LAS format. Oblique photography uses optical cameras to obtain overlapping images and then generates dense 3D point clouds based on multi-view images using structure from motion and multiple view stereo algorithms [18]. Both LiDAR devices and optical cameras can be installed on unmanned aerial vehicles (UAVs) to collect information about a built environment by aerial surveys.…”

Section: Introductionmentioning

confidence: 99%

Automated Simulation Framework for Urban Wind Environments Based on Aerial Point Clouds and Deep Learning

et al. 2021

View full text Add to dashboard Cite

Computational fluid dynamics (CFD) simulation is a core component of wind engineering assessment for urban planning and architecture. CFD simulations require clean and low-complexity models. Existing modeling methods rely on static data from geographic information systems along with manual efforts. They are extraordinarily time-consuming and have difficulties accurately incorporating the up-to-date information of a target area into the flow model. This paper proposes an automated simulation framework with superior modeling efficiency and accuracy. The framework adopts aerial point clouds and an integrated two-dimensional and three-dimensional (3D) deep learning technique, with four operational modules: data acquisition and preprocessing, point cloud segmentation based on deep learning, geometric 3D reconstruction, and CFD simulation. The advantages of the framework are demonstrated through a case study of a local area in Shenzhen, China.

show abstract

Three-dimensional (3D) reconstruction of structures and landscapes: A new point-and-line fusion method

Cited by 19 publications

References 40 publications

3D Indoor Scene Reconstruction and Layout Based on Virtual Reality Technology and Few-Shot Learning

3D Indoor Scene Reconstruction and Layout Based on Virtual Reality Technology and Few-Shot Learning

True Orthophoto Generation Based on Unmanned Aerial Vehicle Images Using Reconstructed Edge Points

Automated Simulation Framework for Urban Wind Environments Based on Aerial Point Clouds and Deep Learning

Contact Info

Product

Resources

About