scholarly journals Making Japenese Ukiyo-e Art 3D in Real-Time

Sci ◽  
2020 ◽  
Vol 2 (2) ◽  
pp. 32
Author(s):  
Innes Brown ◽  
Ognjen Arandjelović
Keyword(s):  
Computer Graphics ◽  
3D Models ◽  
Perspective Projection ◽  
Additional Information ◽  
Interactive 3D ◽  
Line Work ◽  
Japanese Painting ◽  
Distinct Line ◽  
2D Images ◽  
Rendering Techniques

Ukiyo-e is a traditional Japanese painting style most commonly printed using wood blocks. Ukiyo-e prints feature distinct line work, bright colours, and a non-perspective projection. Most previous research on ukiyo-e styled computer graphics has been focused on creation of 2D images. In this paper we propose a framework for rendering interactive 3D scenes with ukiyo-e style. The rendering techniques use standard 3D models as input and require minimal additional information to automatically render scenes in a ukiyo-e style. The described techniques are evaluated based on their ability to emulate ukiyo-e prints, performance, and temporal coherence.

scholarly journals Making Japenese Ukiyo-e Art 3D in Real-Time

Sci ◽  
2020 ◽  
Vol 2 (1) ◽  
pp. 6 ◽  
Author(s):  
Innes Brown ◽  
Ognjen Arandjelović
Keyword(s):  
Computer Graphics ◽  
3D Models ◽  
Perspective Projection ◽  
Additional Information ◽  
Interactive 3D ◽  
Line Work ◽  
Japanese Painting ◽  
Distinct Line ◽  
2D Images ◽  
Rendering Techniques

Ukiyo-e is a traditional Japanese painting style most commonly printed using wood blocks. Ukiyo-e prints feature distinct line work, bright colours, and a non-perspective projection. Most previous research on ukiyo-e styled computer graphics has been focused on creation of 2D images. In this paper we propose a framework for rendering interactive 3D scenes with ukiyo-e style. The rendering techniques use standard 3D models as input and require minimal additional information to automatically render scenes in a ukiyo-e style. The described techniques are evaluated based on their ability to emulate ukiyo-e prints, performance, and temporal coherence.

Automatic generation of bas-relief on 3D models based on 2D images for rhinoceros

2018 ◽  
Vol 23 (6) ◽  
pp. 99-113
Author(s):  
Sha LIU ◽  
Feng YANG ◽  
Shunxi WANG ◽  
Yu CHEN
Keyword(s):  
3D Models ◽  
Automatic Generation ◽  
2D Images

Structure-from-Motion, Capturing Geological Samples v1

2021 ◽  
Author(s):  
Madalyn Massey
Keyword(s):  
United States ◽  
Structure From Motion ◽  
3D Models ◽  
Geological Survey ◽  
United States Geological Survey ◽  
Geological Samples ◽  
Motion Capturing ◽  
2D Images

Structure-from-Motion (SfM) is a photogrammetry process that creates 3D models from overlapping 2D images. This protocol focuses on its application related to geological and geophysical samples. The samples includes fossil, hand samples and rocks. This is a recommended practice to be used later for the publication on United States Geological Survey website.

Telemetry-Based Depth Recovery for Virtual Factory Construction and Extension to Remote Facility Management

1999 ◽  
Author(s):  
Dan Zetu ◽  
Pat Banerjee ◽  
Ali Akgunduz
Keyword(s):  
3D Models ◽  
Human Motion ◽  
Facility Management ◽  
Virtual Manufacturing ◽  
Shop Floor ◽  
Virtual Factory ◽  
Depth Recovery ◽  
Camera Location ◽  
Virtual Manufacturing Environment ◽  
2D Images

Abstract The fast construction of a Virtual Factory model without using a CAD package can be made possible by using computer vision techniques. In order to create a realistic Virtual Manufacturing environment, especially when such a model has to be created in correlation to an existing facility, a reliable algorithm that extracts 3D models from camera images is needed, and this requires exact knowledge of the camera location when capturing images. In this paper, we describe an approach for depth recovery from 2D images based on tracking a camera within the environment. We also explore the extension of our telemetry-based algorithm to remote facility management, by tracking and synchronizing human motion on the shop floor with motion of an avatar in a Virtual Environment representing the same shop floor.

Automatic Creation of Virtual Artefacts from Video Sequences

2005 ◽  
Author(s):  
Andrew W. Fitzgibbon ◽  
Geoff Cross ◽  
Andrew Zisserman
Keyword(s):  
Focal Length ◽  
3D Models ◽  
Video Sequences ◽  
Digital Representation ◽  
The Third ◽  
3D Presentation ◽  
Object Features ◽  
The Given ◽  
System Geometry ◽  
2D Images

Digital representation of an artefact is necessary in order to measure, admire and analyse such ancient pieces. For the purpose of storing, recoding and transmitting information, digital photographs may be enough. However, in the examination purposes of an artefact, a 3D presentation is invaluable as it allows the object viewpoint to be modified freely and 3D measurements to be taken on object features. This chapter describes the system by which 3D models from photographs can be acquired, without the need for the calibration of system geometry such as the camera focal length, relative motion of the camera and object, and the relative positions of the camera and object. This system instead computes the representation of all possible objects and camera configurations which are consistent with the given image. The first section discusses how tracking points observed in 2D images allows for the computation of the relative camera and object geometry. The second section discusses the construction of a triangulated 3D model from the object projections. The third section discusses the refinement of the model based on surface texture.

Classification of Left Atrial Appendage Morphology Using Deep Learning

2020 ◽  
Author(s):  
Mikayle A. Holm ◽  
Alex Deakyne ◽  
Erik Gaasedelen ◽  
Weston Upchurch ◽  
Paul A. Iaizzo
Keyword(s):  
Left Atrial ◽  
Left Atrial Appendage ◽  
Blood Clot ◽  
3D Models ◽  
Atrial Appendage ◽  
Blood Clots ◽  
Mri Scans ◽  
2D Images ◽  
Trained Network

Abstract Atrial fibrillation, a common cardiac arrhythmia, can lead to blood clots in the left atrial appendage (LAA) of the heart, increasing the risk of stroke. Understanding the LAA morphology can indicate the likelihood of a blood clot. Therefore, a classification convolutional neural network was implemented to predict the LAA morphology. Using 2D images of 3D models created from MRI scans of fixed human hearts and a pre-trained network, an 8.7% error rate was achieved. The network can be improved with more data or expanded to classify the LAA from the automatically segmented DICOM datasets and measure the LAA ostia.

Sign in / Sign up

Export Citation Format

Share Document