Learning Hybrid Semantic Affinity for Point Cloud Segmentation

Song, Zhanjie; Zhao, Linqing; Zhou, Jie

doi:10.1109/tcsvt.2021.3132047

Cited by 33 publications

(9 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…LASzip (Isenburg, 2013) is a lossless, order-preserving codec specialized for LiDAR point clouds treating them as sequences and using metadata related to the LiDAR scanning mechanism (ASPRS, 2019). In the special case of spinning LiDAR sensors, Song et al (2021) separate the point cloud into ground, object and noise layers. Typically, most points belong to the ground, and they form circular patterns which can be efficiently compressed.…”

Section: Lidar Point Cloudsmentioning

confidence: 99%

Survey on Deep Learning-Based Point Cloud Compression

Quach¹,

Pang²,

Tian³

et al. 2022

Front. Signal Process.

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

Section: Lidar Point Cloudsmentioning

confidence: 99%

Survey on Deep Learning-Based Point Cloud Compression

Quach¹,

Pang²,

Tian³

et al. 2022

Front. Signal Process.

View full text Add to dashboard Cite

show abstract

“…s in the main paper), nsample specifies the number of points in a local area (i.e. n in the main paper), the parameter [32,128] means that there are two hidden layers in the PSNet network, and the numbers of channels in these two layers are 32 and 128, respectively. Note that the numbers of channels in the input and output layers of the PSNet network are 5 (5 input spatial features) and s, respectively.…”

Section: Channels Of Mlpsmentioning

confidence: 99%

PSNet: Fast Data Structuring for Hierarchical Deep Learning on Point Cloud

Gao

2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…9 visualizes the results of part segmentation initialized with our model, showing that our method can achieve a excellent performance for part segmentation. 3) Semantic segmentation: Semantic segmentation, a technique that associates points or voxels with semantic object labels [58], is a fundamental research challenge in point cloud processing. We use this task to evaluate the effectiveness of our method on data that goes beyond simple, free-standing objects.…”

Section: A Implementation Detailsmentioning

confidence: 99%

Unsupervised Learning on 3D Point Clouds by Clustering and Contrasting

Mei¹,

Yu²,

Wu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Learning from unlabeled or partially labeled data to alleviate human labeling remains a challenging research topic in 3D modeling. Along this line, unsupervised representation learning is a promising direction to auto-extract features without human intervention. This paper proposes a general unsupervised approach, named ConClu, to perform the learning of point-wise and global features by jointly leveraging point-level clustering and instance-level contrasting. Specifically, for one thing, we design an Expectation-Maximization (EM) like soft clustering algorithm that provides local supervision to extract discriminating local features based on optimal transport. We show that this criterion extends standard cross-entropy minimization to an optimal transport problem, which we solve efficiently using a fast variant of the Sinkhorn-Knopp algorithm. For another, we provide an instance-level contrasting method to learn the global geometry, which is formulated by maximizing the similarity between two augmentations of one point cloud. Experimental evaluations on downstream applications such as 3D object classification and semantic segmentation demonstrate the effectiveness of our framework and show that it can outperform state-of-the-art techniques.

show abstract

Learning Hybrid Semantic Affinity for Point Cloud Segmentation

Cited by 33 publications

References 69 publications

Survey on Deep Learning-Based Point Cloud Compression

Survey on Deep Learning-Based Point Cloud Compression

PSNet: Fast Data Structuring for Hierarchical Deep Learning on Point Cloud

Unsupervised Learning on 3D Point Clouds by Clustering and Contrasting

Contact Info

Product

Resources

About