DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

3D Mesh Segmentation Using Transformer Based Graph Operations

Katam, Harinandan (2021) 3D Mesh Segmentation Using Transformer Based Graph Operations. Master's. DLR-Interner Bericht. DLR-IB-RM-OP-2021-10.

[img] PDF


Deep Learning and especially convolutions have been a massive success in computer vision tasks such as Segmentation, Object Detection, and others. However, all of these are limited to 2D images, whereas the progress in the 3D domain has been limited. Extending these priors works to the 3D domain is sadly not straightforward. The biggest challenge here is the unstructured representation of 3D data such as meshes or point clouds. While other works use voxel grids, which have structured representation, they usually struggle with computation time and memory. In this thesis, the task of extending a convolution operation to unstructured data and its problems along with potential solutions is explored. In this thesis, two new methods to the task of Mesh Segmentation are proposed. Both these methods are based on Transformer networks and their components. In the first method, a first of its kind application of transformers to the task of Mesh Segmentation is proposed. In the second method, a permutation invariant Graph Convolution layer named TransConv is proposed which acts similar to a convolution operation on images and can be used in any model architecture. In addition to these methods, two extensions are proposed that improve the performance of both our methods. The first extension is to use depth encoding to add more information about the geodesic distance to the model. The second extension is to extend the concept of atrous convolutions in images to meshes. All of our methods and extensions are evaluated on two datasets and compared with other related works. The first dataset is a collection of high-resolution meshes called the Coseg dataset. The second dataset is a collection of point clouds of 3D objects called ShapeNet part annotation. Our proposed graph convolution layer TransConv outperforms other related works in both the datasets. However, our method to use transformers for mesh segmentation produced comparable results.

Item URL in elib:https://elib.dlr.de/140566/
Document Type:Monograph (DLR-Interner Bericht, Master's)
Title:3D Mesh Segmentation Using Transformer Based Graph Operations
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Katam, HarinandanHarinandan.Katam (at) dlr.deUNSPECIFIED
Date:January 2021
Refereed publication:No
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
Keywords:Deep Learning, Mesh Segmentation, Semantic Segmentation
Institution:Technische Universität München
Department:Fakultät für Informatik
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Space System Technology
DLR - Research area:Raumfahrt
DLR - Program:R SY - Space System Technology
DLR - Research theme (Project):R - Vorhaben Multisensorielle Weltmodellierung (old)
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Robotics and Mechatronics (since 2013) > Perception and Cognition
Deposited By: Denninger, Maximilian
Deposited On:21 Jan 2021 11:56
Last Modified:21 Jan 2021 11:56

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.