A Novel Point Cloud Compression Algorithm Based on Clustering

Sun, Xuebin; Ma, Han; Sun, Yuxiang; Liu, Ming

doi:10.1109/lra.2019.2900747

Cited by 75 publications

(36 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Since depth data represents the points at 3D space, the encoding distortion is evaluated to the differences of the position at 3D space, instead of pixel differences. In the previous studies [23][24][25] of depth data compression, the error of a reconstructed depth data is evaluated through RMSE. RMSE is calculated as follows:…”

Section: Simulation Resultsmentioning

confidence: 99%

Efficient Depth Data Coding Method Based on Plane Modeling for Intra Prediction

2021

View full text Add to dashboard Cite

In this paper, we propose a depth data coding method by plane modeling. The plane modeling is a prediction method based on a plane estimation from depth pixels in a block. The plane modeling improves the intra-picture prediction for depth data comparing with intra modes of conventional coding standards. The plane modeling coefficients, which are the information about the estimated plane, are required to be provided during the depth data coding. The plane modeling coefficients are predicted from neighboring depth pixels. A prediction error of the plane modeling coefficients is calculated through the plane modeling for the selected pixels. If the prediction error is below a certain threshold, then the plane modeling coefficients are applied to the plane modeling for the block. From the simulation results, we verify that the proposed method achieves up to 6.76% bit rate saving in the same coding distortion condition compared to VVC test model.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Efficient Depth Data Coding Method Based on Plane Modeling for Intra Prediction

2021

View full text Add to dashboard Cite

show abstract

“…The depth image is equivalent to the point cloud, but pixels encode distance or depth co-ordinates. Working with depth images makes neighborhood search solutions simpler to manipulate which can significantly reduce the computation complexity [11].…”

Section: Related Workmentioning

confidence: 99%

Linear feature extraction from point cloud using color information

Alshawabkeh

2020

Herit Sci

View full text Add to dashboard Cite

LiDAR is considered as an effective technology for digitizing the real scene at a very high-resolution and in a short time. However, the resolution of the LiDAR is not sufficient to identify and evaluate the façade surface features like edges and cracks. Generally, photographs provide a better interpretation of the linear characteristics. The complementary benefits of each allow exploring valuable spatial information with different surface detail levels. The paper introduces a flexible image-based approach for linear feature extraction from LiDAR point cloud. Initially, the algorithm converts the point clouds into a structured depth image to reduce the complexity and computation time. Using transformation matrix and camera calibration parameters, the visible point clouds are perceptively projected into color image space using co-linearity equations. The result depth channel is sampled with the interpolation process and added to the color channels to compute (RGBD) layers. The edges and linear features of the surface are initially extracted using the optical 2D imagery and subsequently, each pixel of the linear features can be projected directly into 3D space. Due to the various acquisition positions of the laser and color images, the issue of occlusion is resolved using the visibility algorithm. Applying the methodology, experimental results from the Treasury Monument of Jordan's ancient Petra City, indicate that the developed approach provides adequate contour points for better interpretation and quantification of weathering processes and dangerous cracking. 3D Modeling these features can also reduce data size, facilitating surface inspection and analysis with simpler models.

show abstract

“… (1) Expensive than the visible‐light cameras, but cost lower than LIDAR. (2) Thermal images are greyscale visual images in nature. So, the advanced computer vision technology could directly support applications for thermal imaging. (3) It provides dense images in real time, similar to the visible camera. While LIDAR point clouds have a different type of data samples when compared with images, they are sparse point lists, rather than dense arrays [12, 13 ]. For example, the FLIR automotive thermal cameras could stream thermal images with a resolution of

640 \times 512

and run at 60 Hz.…”

Section: Related Workmentioning

confidence: 99%

ABiFN: Attention‐based bi‐modal fusion network for object detection at night time

Charan¹,

Jitesh²,

Chowdhury³

et al. 2020

Electron. lett.

View full text Add to dashboard Cite

Camera‐based object detection in low‐light/night‐time conditions is a fundamental problem because of insufficient lighting. So far, a mid‐level fusion of RGB and thermal images is done to complement each other's features. In this work, an attention‐based bi‐modal fusion network is proposed for a better object detection in the thermal domain by integrating a channel‐wise attention module. The experimental results show that the proposed framework improves the mAP by 4.13 points on the FLIR dataset.

show abstract

A Novel Point Cloud Compression Algorithm Based on Clustering

Cited by 75 publications

References 24 publications

Efficient Depth Data Coding Method Based on Plane Modeling for Intra Prediction

Efficient Depth Data Coding Method Based on Plane Modeling for Intra Prediction

Linear feature extraction from point cloud using color information

ABiFN: Attention‐based bi‐modal fusion network for object detection at night time

Contact Info

Product

Resources

About