Refining The Bounding Volumes for Lossless Compression of Voxelized Point Clouds Geometry

Kaya, Emre Can; Schwarz, Sebastian; Tăbuş, Ioan

doi:10.1109/icip42928.2021.9506767

Cited by 5 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Structured data formats intuitively present the point cloud data and fit for convolution operations naturally. There are mainly voxel-based (Limuti, Polo, and Milani 2018;Quach, Valenzise, and Dufaux 2019;Kaya, Schwarz, and Tabus 2021) and image-based methods (Houshiar and Nüchter 2015;Sun et al 2019;Feng, Liu, and Zhu 2020). The video point cloud compression standard (VPCC) (Schwarz et al 2018) converts the 3D input point cloud into a set of 2D patches followed by a packing process.…”

Section: Related Work Structured Point Cloud Compressionmentioning

confidence: 99%

OctFormer: Efficient Octree-Based Transformer for Point Cloud Compression with Local Enhancement

Cui

Long

Feng

et al. 2023

AAAI

View full text Add to dashboard Cite

Point cloud compression with a higher compression ratio and tiny loss is essential for efficient data transportation. However, previous methods that depend on 3D convolution or frequent multi-head self-attention operations bring huge computations. To address this problem, we propose an octree-based Transformer compression method called OctFormer, which does not rely on the occupancy information of sibling nodes. Our method uses non-overlapped context windows to construct octree node sequences and share the result of a multi-head self-attention operation among a sequence of nodes. Besides, we introduce a locally-enhance module for exploiting the sibling features and a positional encoding generator for enhancing the translation invariance of the octree node sequence. Compared to the previous state-of-the-art works, our method obtains up to 17% Bpp savings compared to the voxel-context-based baseline and saves an overall 99% coding time compared to the attention-based baseline.

show abstract

Section: Related Work Structured Point Cloud Compressionmentioning

confidence: 99%

OctFormer: Efficient Octree-Based Transformer for Point Cloud Compression with Local Enhancement

Cui

Long

Feng

et al. 2023

AAAI

View full text Add to dashboard Cite

show abstract

“…The scientific literature has witnessed a strong interest in improving the compression performance of G-PCC, with many scholarly contributions in recent years, e.g., [6]- [22]. These methods differ in several aspects, such as the representation used for the point cloud (e.g., octree [15]- [20], dyadic decomposition [6], [7], [12] or projections onto 2D planes [13], [23]), the selection of the context used for the conditional probability model for arithmetic coding and the method to define the symbols to be encoded.…”

Section: Introductionmentioning

confidence: 99%

Lossless Compression of Point Cloud Sequences Using Sequence Optimized CNN Models

Kaya

Tăbuş

2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

In this paper we propose a new paradigm for encoding the geometry of dense point cloud sequences, where a convolutional neural network (CNN), which estimates the encoding distributions, is optimized on several frames of the sequence to be compressed. We adopt lightweight CNN structures, we perform training as part of the encoding process and the CNN parameters are transmitted as part of the bitstream. The newly proposed encoding scheme operates on the octree representation for each point cloud, consecutively encoding each octree resolution level. At every octree resolution level, the voxel grid is traversed section-by-section (each section being perpendicular to a selected coordinate axis), and in each section, the occupancies of groups of two-by-two voxels are encoded at once in a single arithmetic coding operation. A context for the conditional encoding distribution is defined for each two-by-two group of voxels based on the information available about the occupancy of the neighboring voxels in the current and lower resolution layers of the octree. The CNN estimates the probability mass functions of the occupancy patterns of all the voxel groups from one section in four phases. In each new phase, the contexts are updated with the occupancies encoded in the previous phase, and each phase estimates the probabilities in parallel, providing a reasonable trade-off between the parallelism of the processing and the informativeness of the contexts. The CNN training time is comparable to the time spent in the remaining encoding steps, leading to competitive overall encoding times. The bitrates and encoding-decoding times compare favorably with those of recently published compression schemes.INDEX TERMS Convolutional neural networks, lossless geometry compression, octree coding, point cloud compression

show abstract

A Du-Octree based Cross-Attention Model for LiDAR Geometry Compression

Cui,

Feng,

Long

et al. 2024

2024 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Refining The Bounding Volumes for Lossless Compression of Voxelized Point Clouds Geometry

Cited by 5 publications

References 13 publications

OctFormer: Efficient Octree-Based Transformer for Point Cloud Compression with Local Enhancement

OctFormer: Efficient Octree-Based Transformer for Point Cloud Compression with Local Enhancement

Lossless Compression of Point Cloud Sequences Using Sequence Optimized CNN Models

A Du-Octree based Cross-Attention Model for LiDAR Geometry Compression

Contact Info

Product

Resources

About