Towards Efficient Point Cloud Graph Neural Networks Through Architectural Simplification

Tailor, Shyam A.; Jong, René de; Azevedo, Tiago; Mattina, Matthew; Maji, Partha

doi:10.1109/iccvw54120.2021.00237

Cited by 9 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Geometric Relation Supplement Spatial geometric information is critical for model performance [40]. In order to provide more geometric information, we transform coordinates at each scale and their neighbours into the spherical coordinate system to obtain relative angles for more spatial relation description…”

Section: Geometric Space Learningmentioning

confidence: 99%

See 1 more Smart Citation

PointGS: Bridging and fusing geometric and semantic space for 3D point cloud analysis

Jiang

Huang²,

Wu³

et al. 2023

Information Fusion

View full text Add to dashboard Cite

Section: Geometric Space Learningmentioning

confidence: 99%

“…According to [40], it is feasible to simplify the local features updating rule with negligible performance drop. They present a simple version to distill edges:…”

Section: Semantic Space Learningmentioning

confidence: 99%

PointGS: Bridging and fusing geometric and semantic space for 3D point cloud analysis

Jiang

Huang²,

Wu³

et al. 2023

Information Fusion

View full text Add to dashboard Cite

“…Baruah et al [13] introduce GNNMark, a benchmarking suite for GNNs on GPUs. Tailor et al [193] analyze the performance of point cloud GNNs. Qiu et al [171] analyze the impact of sparse matrix data format on the performance of GNN computations.…”

Section: Analyses and Evaluations Of Gnn Systemsmentioning

confidence: 99%

Parallel and Distributed Graph Neural Networks: An In-Depth Concurrency Analysis

Besta¹,

Hoefler²

2022

Preprint

View full text Add to dashboard Cite

Graph neural networks (GNNs) are among the most powerful tools in deep learning. They routinely solve complex problems on unstructured networks, such as node classification, graph classification, or link prediction, with high accuracy. However, both inference and training of GNNs are complex, and they uniquely combine the features of irregular graph processing with dense and regular computations. This complexity makes it very challenging to execute GNNs efficiently on modern massively parallel architectures. To alleviate this, we first design a taxonomy of parallelism in GNNs, considering data and model parallelism, and different forms of pipelining. Then, we use this taxonomy to investigate the amount of parallelism in numerous GNN models, GNN-driven machine learning tasks, software frameworks, or hardware accelerators. We use the work-depth model, and we also assess communication volume and synchronization. We specifically focus on the sparsity/density of the associated tensors, in order to understand how to effectively apply techniques such as vectorization. We also formally analyze GNN pipelining, and we generalize the established Message-Passing class of GNN models to cover arbitrary pipeline depths, facilitating future optimizations. Finally, we investigate different forms of asynchronicity, navigating the path for future asynchronous parallel GNN pipelines. The outcomes of our analysis are synthesized in a set of insights that help to maximize GNN performance, and a comprehensive list of challenges and opportunities for further research into efficient GNN computations. Our work will help to advance the design of future GNNs.

show abstract

“…Recently, studies on the analysis of graphs using machine learning have been gaining more attention because of the outstanding expressive power of graphs. For instance, graphs can be used to denote a large number of applications across different scopes, including the extraction of topologies and geometries from 3D object detection or point cloud [12]- [14], natural science problems [15], [16], pharmaceutical research [17], [18] and other areas [19]. Graph computation, which is a unique form of data structure for supervised and unsupervised learning strategies, focuses on tasks, such as clustering, link prediction, and classification.…”

Section: Introductionmentioning

confidence: 99%

An efficient activity recognition for homecare robots from multi-modal communication dataset

Yani

Yamada

Siow

et al. 2023

Int. J. Adv. Intell. Informatics

View full text Add to dashboard Cite

Human environments are designed and managed by humans for humans. Thus, adding robots to interact with humans and perform specific tasks appropriately is an essential topic in robotics research. In recent decades, object recognition, human skeletal, and face recognition frameworks have been implemented to support the tasks of robots. However, recognition of activities and interactions between humans and surrounding objects is an ongoing and more challenging problem. Therefore, this study proposed a graph neural network (GNN) approach to directly recognize human activity at home using vision and speech teaching data. Focus was given to the problem of classifying three activities, namely, eating, working, and reading, where these activities were conducted in the same environment. From the experiments, observations, and analyses, this proved to be quite a challenging problem to solve using only traditional convolutional neural networks (CNN) and video datasets. In the proposed method, an activity classification was learned from a 3D detected object corresponding to the human position. Next, human utterances were used to label the activity from the collected human and object 3D positions. The experiment, involving data collection and learning, was demonstrated by using human-robot communication. It was shown that the proposed method had the shortest training time of 100.346 seconds with 6000 positions from the dataset and was able to recognize the three activities more accurately than the deep layer aggregation (DLA) and X3D networks with video datasets.

show abstract

Towards Efficient Point Cloud Graph Neural Networks Through Architectural Simplification

Cited by 9 publications

References 18 publications

PointGS: Bridging and fusing geometric and semantic space for 3D point cloud analysis

PointGS: Bridging and fusing geometric and semantic space for 3D point cloud analysis

Parallel and Distributed Graph Neural Networks: An In-Depth Concurrency Analysis

An efficient activity recognition for homecare robots from multi-modal communication dataset

Contact Info

Product

Resources

About