Lossy Point Cloud Geometry Compression via Region-Wise Processing

Zhu, Wen-Hong; Xu, Yiling; Ding, Dandan; Ma, Zhan; Nilsson, Mike

doi:10.1109/tcsvt.2021.3101852

Cited by 20 publications

(6 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, a point cloud autoencoder network framework with quantization layers is proposed to learn compact latent feature representations from each sub-block. Subsequently, Zhu et al [ 30 ] exploited regional similarity to achieve efficient lossy point cloud geometry compression. To this end, the input point cloud is divided into multiple local regions, and they are grouped into distinct clusters according to the region surface vectors, ensuring that the inter-cluster similarity is minimized and the intra-cluster similarity is maximized.…”

Section: Related Workmentioning

confidence: 99%

Point Cloud Geometry Compression Based on Multi-Layer Residual Structure

Wang

Sun

et al. 2022

Entropy

View full text Add to dashboard Cite

Point cloud data are extensively used in various applications, such as autonomous driving and augmented reality since it can provide both detailed and realistic depictions of 3D scenes or objects. Meanwhile, 3D point clouds generally occupy a large amount of storage space that is a big burden for efficient communication. However, it is difficult to efficiently compress such sparse, disordered, non-uniform and high dimensional data. Therefore, this work proposes a novel deep-learning framework for point cloud geometric compression based on an autoencoder architecture. Specifically, a multi-layer residual module is designed on a sparse convolution-based autoencoders that progressively down-samples the input point clouds and reconstructs the point clouds in a hierarchically way. It effectively constrains the accuracy of the sampling process at the encoder side, which significantly preserves the feature information with a decrease in the data volume. Compared with the state-of-the-art geometry-based point cloud compression (G-PCC) schemes, our approach obtains more than 70–90% BD-Rate gain on an object point cloud dataset and achieves a better point cloud reconstruction quality. Additionally, compared to the state-of-the-art PCGCv2, we achieve an average gain of about 10% in BD-Rate.

show abstract

Section: Related Workmentioning

confidence: 99%

Point Cloud Geometry Compression Based on Multi-Layer Residual Structure

Wang

Sun

et al. 2022

Entropy

View full text Add to dashboard Cite

show abstract

“…With the increasing capability of 3D acquisition device, it becomes critical that how to efficiently compress 3D point clouds. Currently, PCG compression methods are mainly focused on octree based solutions [8,14,16,24,35], 3D convolutional autoencoder [22,23,27,28] and PointNet-based autoencoder [29,31,32].…”

Section: Related Workmentioning

confidence: 99%

“…This algorithm is easy to understand and implement, but it cannot perform well at low bit rate, and the number of points generated will decrease sharply with the decrease of tree depth. The authors of [35] uses a region-wise processing to extract information redundancy between point cloud surface regions, which can reduce the number of bits required for compression to a certain extent. Some methods also combine deep learning with octree method to pursue better performance, such as [8,24].…”

Section: Related Workmentioning

confidence: 99%

“…The 3D point cloud is an important representation of objects or scenes in 3D space [1,7,10,34]. The highly challenging task of compressing the shape information represented by a set of coordinates of points in a point cloud is referred to as point cloud geometry (PCG) compression [11,25,27,35]. With the rapid development of point cloud applications in recent years, PCG compression is gaining more and more attention.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Patch-Based Deep Autoencoder for Point Cloud Geometry Compression

You

Gao

2021

ACM Multimedia Asia

View full text Add to dashboard Cite

The ever-increasing 3D application makes the point cloud compression unprecedentedly important and needed. In this paper, we propose a patch-based compression process using deep learning, focusing on the lossy point cloud geometry compression. Unlike existing point cloud compression networks, which apply feature extraction and reconstruction on the entire point cloud, we divide the point cloud into patches and compress each patch independently. In the decoding process, we finally assemble the decompressed patches into a complete point cloud. In addition, we train our network by a patch-to-patch criterion, i.e., use the local reconstruction loss for optimization, to approximate the global reconstruction optimality. Our method outperforms the state-of-the-art in terms of rate-distortion performance, especially at low bitrates. Moreover, the compression process we proposed can guarantee to generate the same number of points as the input. The network model of this method can be easily applied to other point cloud reconstruction problems, such as upsampling. CCS CONCEPTS• Computing methodologies → Reconstruction; • Theory of computation → Data compression.

show abstract

“…In recent years, 3D point clouds have been increasingly applied in many fields, such as autonomous driving, history heritage restoration, virtual reality (AR/VR), and immersive communication [1][2][3]. A 3D point cloud usually consists of a set of points that represent the positions (x, y, z) and attributes (such as color and reflectance) of objects in 3D space [4][5].…”

Section: Introductionmentioning

confidence: 99%

Graph Filter-Based Fast Motion Matching for Inter Frame Coding of MPEG G-PCC

Guo

Yuan

Wang

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

As a key technique for geometry-based point cloud inter frame coding, motion estimation (ME) has shown the effectiveness to improve the coding efficiency. But the high complexity limits its application. We propose a novel and effective method for fast global motion matching between consecutive frames. We first divide an input point cloud into several blocks, and construct a graph for each block. Then, a graph filter is applied to remove the highest and lowest frequency coefficients based on predefined thresholds. Finally, by quantitatively analyzing the geometry gradients of the output of the graph filter, only some key points are reserved. Experimental results show that an average 49% encoding time can be saved with only 0.5% loss of coding efficiency for lossy geometry-lossy attributes configuration while an average 45% encoding time can be saved with no loss of coding efficiency for lossless geometry-lossless attributes configuration.

show abstract

Lossy Point Cloud Geometry Compression via Region-Wise Processing

Cited by 20 publications

References 42 publications

Point Cloud Geometry Compression Based on Multi-Layer Residual Structure

Point Cloud Geometry Compression Based on Multi-Layer Residual Structure

Patch-Based Deep Autoencoder for Point Cloud Geometry Compression

Graph Filter-Based Fast Motion Matching for Inter Frame Coding of MPEG G-PCC

Contact Info

Product

Resources

About