Interactive shape co-segmentation via label propagation

By Zizhao Wu, Ruyang Shou, Yunhai Wang, Xinguo Liu. 2014

48 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition

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

1. Leonidas J. Guibas
2. Charles Ruizhongtai Qi
3. Li Yi
4. Hao Su
5. Or Litany
6. Matthias Nießner
7. Vladimir G. Kim
8. Andreas Geiger

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Xiaogang Wang
3. Leonidas J. Guibas
4. Jitendra Malik
5. Dacheng Tao
6. Ming-Hsuan Yang
7. Trevor Darrell
8. Andrea Vedaldi

Citations in the corpus

• PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (2016, methodology): “In this section, we compare our segmentation version PointNet (a modified version of Fig 2, Segmentation Network) with two traditional methods [24] and [26] that both take advantage of point-wise geometry features and correspondences between shapes, as well as our own 3D CNN baseline.”
• SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation (2016, neutral): “We compare our framework with traditional learning-based techniques [27, 29] leveraging on local geometric features and shape alignment cues, as well as recent deep learning based approaches [2] which also fall into the family of graph CNNs.”
• FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis (2017, methodology): “We follow the standard experimental protocol [12, 19, 31, 33], and report the mean intersection over union (mIoU) metric per category and across all shapes.”
• A scalable active framework for region annotation in 3D shape collections (2016, methodology): “Previous techniques that only rely on point-wise shape features [Wu et al. 2014] are sensitive to outliers and large shape variations often observed in massive datasets.”
• Attentional ShapeContextNet for Point Cloud Recognition (2018, result): “We compared the results with Wu [37], Yi [42], 3DCNN from [6], PointNet [6] and recent PointNet++[25] which uses additional normal direction features.”
• PointGrid: A Deep Network for 3D Shape Understanding (2018, methodology): “Experiments on widely used benchmark datasets [57, 7, 62] show that PointGrid compares favorably over existing deep learning methods [56, 62, 38, 27, 63, 55] on both classification and segmentation with significantly smaller memory footprint than other volumetric approaches.”
• Deep Learning on Point Sets for 3 D Classification and Segmentation (2016, methodology): “In this section, we compare our segmentation version PointNet (Fig 2, Segmentation Network) with two traditional methods [24] and [26] that both take advantage of point-wise geometric features and correspondences between shapes, as well as our volumetric deep learning baseline (3DCNN).”