On using PointNet Architecture for Human Body Segmentation

• Published in: 2019 11th International Symposium on Image and Signal Processing and Analysis (ISPA) pp. 253 - 257
• Main Authors: Jertec, Andrej, Bojanic, David, Bartol, Kristijan, Pribanic, Tomislav, Petkovic, Tomislav, Petrak, Slavenka
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2019
• Subjects: 3D shape analysis; Biological system modeling; Data models; deep learning; human body segmentation; Image segmentation; PointNet; Shape; Solid modeling; Three-dimensional displays; Training
• ISSN: 1849-2266
• DOI: 10.1109/ISPA.2019.8868844

In the case of structured data, such as 2D images, many variants of traditional convolution neural network architectures have been successfully proposed. Learning from unstructured sets of data, such as sets of 3D point clouds, is a challenging task due to numerous reasons among which two most important ones are: 3D point cloud is generally (i) unordered and (ii) sparse data set. Therefore, the architectures have been proposed which are invariant to both ordering and number of points in the point cloud. PointNet is one such architecture, originally introduced and demonstrated on the task of classification and segmentation of the ModelNet40 data set. In this work we study the performance of PointNet on an even more demanding task, segmentation of human body parts. Finding enough training data of enough quality is generally a problem in deep learning, and especially for human body segmentation. To that end we take advantage of SMPL model which provides human body models in many shapes and sizes in an essentially automatic fashion, therefore avoiding a cumbersome procedure of manual collection and preparation of training data. Our results show that the proposed PointNet variant trained using SMPL model provides competitive segmentation results on the task of human body segmentation.