GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition

By Yifan Feng, Zizhao Zhang, Xibin Zhao, Rongrong Ji, Yue Gao. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018

623 citations recorded by the source corpus.

3D shape recognition has attracted much attention recently. Its recent advances advocate the usage of deep features and achieve the state-of-the-art performance. However, existing deep features for 3D shape recognition are restricted to a view-to-shape setting, which learns the shape descriptor from the view-level feature directly. Despite the exciting progress on view-based 3D shape description, the intrinsic hierarchical correlation and discriminability among views have not been well exploited, which is important for 3D shape representation. To tackle this issue, in this paper, we propose a group-view convolutional neural network (GVCNN) framework for hierarchical correlation modeling towards discriminative 3D shape description. The proposed GVCNN framework is composed of a hierarchical view-group-shape architecture, i.e., from the view level, the group level and the shape level, which are organized using a grouping strategy. Concretely, we first use an expanded CNN to extract a view level descriptor. Then, a grouping module is introduced to estimate the content discrimination of each view, based on which all views can be splitted into different groups according to their discriminative level. A group level description can be further generated by pooling from view descriptors. Finally, all group level descriptors are combined into the shape level descriptor according to their discriminative weights. Experimental results and comparison with state-of-the-art methods show that our proposed GVCNN method can achieve a significant performance gain on both the 3D shape classification and retrieval tasks.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Geology

View Paper: https://www.semanticscholar.org/paper/f60baea800d88a26f7224af76849cb117f3384e3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Geology

Citations in the corpus

Towards 3D VR-Sketch to 3D Shape Retrieval (2020, methodology): “Various input modalities have been tried – images [14] and rough 2D sketch [32] at ﬁrst, with latest research focusing on 3D to 3D, i.e., using 3D scans [1, 5] or existing 3D shapes [30, 24, 21, 7, 9, 10, 12, 23, 25, 31].”
View-GCN: View-Based Graph Convolutional Network for 3D Shape Analysis (2020, neutral): “…++ [42] Points − 91.9 Kd-Networks [30] 88.5 91.8 RS-CNN [33] − 93.6 MVCNN [47] Images 90.1 90.1 MVCNN-new [48] 92.4 95.0 MHBN [55] 93.1 94.7 GVCNN [15] 90.7 93.1 RCPCNN [50] − 93.8 3D2SeqViews[19] 91.5 93.4 SeqViews2SeqLabels [20] 91 signiﬁcantly higher accuracies by more than 3.4% per class and…”
Group-Free 3D Object Detection via Transformers (2021, background): “Network architecture for Point Cloud A large set of network architectures [38, 12, 29, 9, 47, 23, 44, 39, 45, 28, 30, 33, 43, 40, 17, 1, 41, 49, 10, 22] have been proposed for various point cloud based learning tasks.”
PVNet: A Joint Convolutional Network of Point Cloud and Multi-View for 3D Shape Recognition (2018, methodology): “In [6], a group-based method is proposed to exploit the relationship among different views.”
Hypergraph Neural Networks (2018, background): “When more features, i.e., both GVCNN and MVCNN, are used for graph/hypergraph structure generation, HGNN achieves much better performance compared with GCN.”
HGNN+: General Hypergraph Neural Networks (2022, methodology): “Here, two public 3D object datasets are employed, including the ModelNet40 [40] dataset and the NTU [41] [42]) and Group-View Convolutional Neural Network (GVCNN [43]).”
Deep multi-view learning methods: A review (2021, methodology): “For the deep multi-view object recognition methods, we compare the classification accuracy (Accuracy) [49] of the following 660 approaches: multi-view CNN (MVCNN) [49], Kd-tree based network (KD-Net) [51], multi-view harmonized bilinear network (MHBN) [52], group-view CNN (GVCNN) [50] and dynamic ro”
MeshNet: Mesh Neural Network for 3D Shape Representation (2018, background): “Group-view convolutional neural networks (GVCNN) (Feng et al. 2018) proposes a hierarchical framework, which divides views into different groups with different weights to generate a more discriminative descriptor for a 3D shape.”
Deep Multi-View Enhancement Hashing for Image Retrieval (2020, methodology): “Other than this, the multi-view is also well known as multiple angles cameras of 3D model [16].”
PointCLIP: Point Cloud Understanding by CLIP (2021, methodology): “Therein, multiview methods project point clouds onto images of multiple views and process them with 2D Convolution Neural Networks (CNN) [22] pre-trained on ImageNet [28], such as MVCNN [49] and others [14, 15, 21, 26, 62].”
Relation-Shape Convolutional Neural Network for Point Cloud Analysis (2019, background): “Recently, many works [35, 2, 5, 43, 6, 25] have been proposed to recognize these view images with deep neural networks.”
DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processing (2019, methodology): “View-based methods [44, 3, 6, 54, 7, 33, 14] represent a 3D shape as a collection of 2D views, over which classic CNN used in image analysis field can be easily applied.”
Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline (2021, background): “, 2018), GVCNN which groups features from subsets of views together before aggregating into a final prediction (Feng et al., 2018), and hypergraph methods that consider the correlation across training samples (Zhang et al.”