Oregon State University

We’d like your feedback: Calendar User Survey – Event Creator Survey



Event Details

MS Final Examination – Xu Xu

Tuesday, June 13, 2017 3:00 PM - 5:00 PM

Reasoning about 3D Shapes of objects and Synthesis of 3D Shapes from 2D images via Deep Neural Network
The topic of 3D shape reasoning has gained more and more attention in recent few years in 3D object recognition and generation and computer vision. Usually, 3D objects can be recognized based on their shapes, since they are important properties of objects. Recent work typically represents a 3D shape as a set of binary variables corresponding to 3D voxels of a uniform 3D grid centered on the shape, and resorts to deep convolutional neural networks (CNNs) for modeling these binary variables. However, robust learning of such CNNs is currently limited by the small datasets of 3D shapes available – an order of magnitude smaller than other common datasets in computer vision. Recent related work typically deals with the small training datasets using a number of ad hoc, hand-tuning strategies. To address this issue, we formulate CNN learning as a beam search aimed at identifying an optimal CNN architecture as well as estimating parameters of such an optimal CNN. Additionally, a 3D-VAE-GAN(3D Variational AutoEncoder - Generative Adversarial Network)model has also been proposed to synthesize high-quality 3D objects with 2D images from IKEA dataset as inputs.

Major Advisor: Sinisa Todorovic
Committee: Fuxin Li
Committee: Alan Fern
GCR: Henri Jansen

Kelley Engineering Center (campus map)
Calvin Hughes
1 541 737 3168
Calvin.Hughes at oregonstate.edu
Sch Elect Engr/Comp Sci
This event appears on the following calendars: