Generation and modification of 3D models with deep neural networks

Öngün, Cihan
Artificial intelligence (AI) and particularly deep neural networks (DNN) have become very hot topics in the recent years and they have been shown to be successful in problems such as detection, recognition and segmentation. More recently DNNs have started to be popular in data generation problems by the invention of Generative Adversarial Networks (GAN). Using GANs, various types of data such as audio, image or 3D models could be generated. In this thesis, we aim to propose a system that creates artificial 3D models with given characteristics. For this purpose, we focus on latent modification and generation of 3D point cloud object models with respect to their semantic parts. Different to the existing methods which use separate networks for part generation and assembly, we propose a single end-to-end Autoencoder model that can handle generation and modification of both semantic parts, and global shapes. The proposed method supports part exchange between 3D point cloud models and composition by different parts to form new models by directly editing latent representations. This holistic approach does not need part-based training to learn part representations and does not introduce any extra loss besides the standard reconstruction loss. The experiments demonstrate the robustness of the proposed method with different object categories and varying number of points, rotations and scales. The method can generate new models by integration of generative models such as GANs and VAEs and can work with unannotated point clouds by integration of a segmentation module.


A Review on Interpretable and Explainable Artificial Intelligence in Hydroclimatic Applications
Basagaoglu, Hakan; Chakraborty, Debaditya; Do Lago, Cesar; Gutierrez, Lilianna; ŞAHİNLİ, MEHMET ARİF; Giacomoni, Marcio; Furl, Chad; Mirchi, Ali; Moriasi, Daniel; Şengör, Sema Sevinç (2022-04-01)
This review focuses on the use of Interpretable Artificial Intelligence (IAI) and eXplainable Artificial Intelligence (XAI) models for data imputations and numerical or categorical hydroclimatic predictions from nonlinearly combined multidimensional predictors. The AI models considered in this paper involve Extreme Gradient Boosting, Light Gradient Boosting, Categorical Boosting, Extremely Randomized Trees, and Random Forest. These AI models can transform into XAI models when they are coupled with the expla...
Case studies on the use of neural networks in eutrophication modeling
Karul, C; Soyupak, S; Cilesiz, AF; Akbay, N; Germen, E (2000-10-30)
Artificial neural networks are becoming more and more common to be used in development of prediction models for complex systems as the theory behind them develops and the processing power of computers increase. A three layer Levenberg-Marquardt feedforward learning algorithm was used to model the eutrophication process in three water bodies of Turkey (Keban Dam Reservoir, Mogan and Eymir Lakes). Despite the very complex and peculiar nature of Keban Dam, a relatively good correlation (correlation coefficient...
Evaluating the convergence of high-performance computing with big data, artificial intelligence and cloud computing technologies
Dildar Korkmaz, Yeşim; Eren, Pekin Erhan; Kayabay, Kerem; Department of Information Systems (2023-1-24)
The advancements in High-Performance Computing (HPC), Big Data, Artificial Intelligence (AI), and Cloud Computing technologies have led to a convergence of these fields, resulting in the emergence of significant improvements for a wide range of fields. Identifying the state of development of technology convergence and forecasting promising technology convergence is critical for both academia and industry. That's why technology assessment and forecasting for HPC-Big Data-AI-Cloud Computing convergence is nee...
Learning a partially-observable card game hearts using reinforcement learning
Demirdöver, Buğra Kaan; Alpaslan, Ferda Nur; Department of Computer Engineering (2020)
Artificial intelligence and machine learning are widely popular in many sectors. Oneof them is the gaming industry. With many different scenarios, different types, gamesare perfect for machine learning and artificial intelligence. This study aims to developlearning agents to play the game of Hearts. Hearts is one of the most popular cardgames in the world. It is a trick based, imperfect information card game. In additionto having a huge state space, hearts offers many extra challenges due to the nature ofth...
Scalable high-performance architecture for convolutional ternary neural networks on FPGA
Prost-Boucle, Adrien; Bourge, Alban; Petrot, Frederic; Alemdar, Hande; Caldwell, Nicholas; Leroy, Vincent (2017-09-06)
Thanks to their excellent performances on typical artificial intelligence problems, deep neural networks have drawn a lot of interest lately. However, this comes at the cost of large computational needs and high power consumption. Benefiting from high precision at acceptable hardware cost on these difficult problems is a challenge. To address it, we advocate the use of ternary neural networks (TNN) that, when properly trained, can reach results close to the state of the art using floating-point arithmetic. ...
Citation Formats
C. Öngün, “Generation and modification of 3D models with deep neural networks,” Ph.D. - Doctoral Program, Middle East Technical University, 2021.