Lossless Coding of Point Cloud Geometry Using a Deep Generative Model

Nguyen, Dat Thanh; Quach, Maurice; Valenzise, Giuseppe; Duhamel, Pierre

doi:10.1109/tcsvt.2021.3100279

Cited by 44 publications

(28 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More details regarding the sparse convolutions can be also found in [21]. As in Table 2, our sparse tensor-based solution could offer comparable complexity with the MPEG G-PCC anchor, and only requires just a few percentage of running time of those uniform voxel based methods [10], [26] (see Table 4). In the meantime, our method requires a small amount of storage due to model sharing across scales (see Sec.…”

Section: Sparse Convolutionmentioning

confidence: 95%

“…Lossless Compression of Dense PCGs. Nguyen et al [26] developed a learnt lossless compression method for dense PCG, termed as the VoxelDNN, in which masked 3D CNN is applied for voxel occupancy probability approximation in a uniform voxel representation. The VoxelDNN provided promising compression ratios at the expense of a very long encoding and decoding time duration incurred by the sequential processing.…”

Section: Comparison To Learning-based Solutionsmentioning

confidence: 99%

“…The proposed SparsePCGC also demonstrates the lightweight space and time complexity consumption. For example, its computational runtime for encoding and decoding is comparable to the G-PCC (see Table 2), and reduces the running time by orders of magnitude to recently published VoxelDNN [26] and NNOC [25]. And, applying the model sharing across scales would significantly reduce the space requirement for model storage which is attractive to industrial practitioners.…”

mentioning

confidence: 97%

See 2 more Smart Citations

Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compression

Wang¹,

Ding²,

Li³

et al. 2021

Preprint

View full text Add to dashboard Cite

This study develops a unified Point Cloud Geometry (PCG) compression method through Sparse Tensor Processing (STP) based multiscale representation of voxelized PCG, dubbed as the SparsePCGC. Applying the STP reduces the complexity significantly because it only performs the convolutions centered at Most-Probable Positively-Occupied Voxels (MP-POV). And the multiscale representation facilitates us to compress scale-wise MP-POVs progressively. The overall compression efficiency highly depends on the approximation accuracy of occupancy probability of each MP-POV, e.g., p MP-POV . Thus, we design the Sparse Convolution based Neural Networks (SparseCNN) consisting of sparse convolutions and voxel re-sampling to extensively exploit priors. We then develop the SparseCNN based Occupancy Probability Approximation (SOPA) model to estimate the p MP-POV s in a single-stage manner only using the cross-scale prior or in multi-stage by step-wisely utilizing autoregressive neighbors. Besides, we also suggest the SparseCNN based Local Neighborhood Embedding (SLNE) to characterize the local spatial variations as the feature attribute to improve the SOPA. Our unified approach shows the state-of-art performance in both lossless and lossy compression modes across a variety of datasets including the dense PCGs (8iVFB, Owlii) and the sparse LiDAR PCGs (KITTI, Ford) when compared with the MPEG G-PCC and other popular learning-based compression schemes. Furthermore, the proposed method presents lightweight complexity due to point-wise computation, and tiny storage desire because of model sharing across all scales. We make all materials publicly accessible at https://github.com/NJUVISION/SparsePCGC for reproducible research.

show abstract

Section: Sparse Convolutionmentioning

confidence: 95%

Section: Comparison To Learning-based Solutionsmentioning

confidence: 99%

mentioning

confidence: 97%

See 1 more Smart Citation

Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compression

Wang¹,

Ding²,

Li³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The rate-optimized octree partitioning improves performance on sparser regions. The compression performance is further improved with data augmentation and context extension techniques (Nguyen et al, 2021a). The drawback of such approaches is that the sequential dependency increases complexity significantly.…”

Section: Lossless Compressionmentioning

confidence: 99%

“…MPEG (2021d) performed a comparison in the performance of different state of the art point cloud compression approaches. Overall, CNN based approaches (Guarda et al, 2020a;Quach et al, 2020b;Nguyen et al, 2021a;Wang et al, 2021) outperform the G-PCC codec with the drawback of additional computational complexity. CNN based methods seem to perform better on denser point clouds.…”

Section: Compression Performancementioning

confidence: 99%

Survey on Deep Learning-Based Point Cloud Compression

Quach¹,

Pang²,

Tian³

et al. 2022

Front. Signal Process.

Self Cite

View full text Add to dashboard Cite

Point clouds are becoming essential in key applications with advances in capture technologies leading to large volumes of data. Compression is thus essential for storage and transmission. In this work, the state of the art for geometry and attribute compression methods with a focus on deep learning based approaches is reviewed. The challenges faced when compressing geometry and attributes are considered, with an analysis of the current approaches to address them, their limitations and the relations between deep learning and traditional ones. Current open questions in point cloud compression, existing solutions and perspectives are identified and discussed. Finally, the link between existing point cloud compression research and research problems to relevant areas of adjacent fields, such as rendering in computer graphics, mesh compression and point cloud quality assessment, is highlighted.

show abstract

OctPCGC-Net: Learning Octree-Structured Context Entropy Model for Point Cloud Geometry Compression

Wang,

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Lossless Coding of Point Cloud Geometry Using a Deep Generative Model

Cited by 44 publications

References 23 publications

Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compression

Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compression

Survey on Deep Learning-Based Point Cloud Compression

OctPCGC-Net: Learning Octree-Structured Context Entropy Model for Point Cloud Geometry Compression

Contact Info

Product

Resources

About