Learning from graphically integrated 2D and 3D representations improves retention of neuroanatomy.

We introduce a View-Volume convolutional neural network (VVNet) for inferring the occupancy and semantic labels of a volumetric 3D scene from a single depth image. Our method extracts the detailed geometric features from the input depth image with a 2D view CNN and then projects the features into a 3D volume according to the input depth map via a projection layer. After that, we learn the 3D context information of the scene with a 3D volume CNN for computing the result volumetric occupancy and semantic labels. With combined 2D and 3D representations, the VVNet efficiently reduces the computational cost, enables feature extraction from multi-channel high resolution inputs, and thus significantly improve the result accuracy. We validate our method and demonstrate its efficiency and effectiveness on both synthetic SUNCG and real NYU dataset.

Download Full-text

2D and 3D Representations of Solution Spaces for CO Problems

Computational Science - ICCS 2004 - Lecture Notes in Computer Science ◽

10.1007/978-3-540-24687-9_61 ◽

2004 ◽

pp. 483-490 ◽

Cited By ~ 1

Author(s):

Eugeniusz Nowicki ◽

Czesław Smutnicki

Keyword(s):

2D And 3D ◽

3D Representations

Download Full-text

3D Exploration of the Brainstem in 50-micron Resolution MRI

10.1101/768002 ◽

2019 ◽

Author(s):

N Makris ◽

RJ Rushmore ◽

P Wilson-Braun ◽

G Papadimitriou ◽

I Ng ◽

...

Keyword(s):

High Resolution ◽

Cranial Nerve ◽

Emotional Processing ◽

Neural Systems ◽

Clinical Neuroscience ◽

Vital Importance ◽

Nuclear Masses ◽

3D Exploration ◽

2D And 3D ◽

3D Representations

AbstractThe brainstem, a structure of vital importance in the mammals, is currently becoming a principal focus in cognitive, affective and clinical neuroscience. Midbrain, pontine and medullar structures are the epicenter of conduit, cranial nerve and such integrative functions as consciousness, emotional processing, pain and motivation. In this study, we parcellated the nuclear masses and the principal fiber pathways that were visible in a high resolution T2-weighted MRI dataset of 50-micron isotropic voxels of a postmortem human brainstem. Based on this analysis, we generated a detailed map of the human brainstem. To assess the validity of our maps, we compared our observations with histological maps of traditional human brainstem atlases. Moreover, we reconstructed the motor, sensory and integrative neural systems of the brainstem and rendered them in 3D representations. We anticipate the utilization of these maps by the neuroimaging community at large for applications in basic neuroscience as well as in neurology, psychiatry and neurosurgery, due to their versatile computational nature in 2D and 3D representations in a publicly available capacity.

Download Full-text

A Study of Mobile Guide Applications in Wayfinding Context

Advances in Human and Social Aspects of Technology - Human-Computer Interfaces and Interactivity ◽

10.4018/978-1-4666-6228-5.ch013 ◽

2014 ◽

pp. 230-246

Author(s):

Yu-Horng Chen ◽

Yih-Shyuan Chen

Keyword(s):

Interface Design ◽

Location Awareness ◽

Research Projects ◽

On Demand ◽

3D Information ◽

Technical Solutions ◽

2D And 3D ◽

Small Screen Devices ◽

3D Representations ◽

Small Screen

This chapter investigates the research projects and prototypes related to currently available mobile applications used in wayfinding and navigation. The advancement in small-screen devices, mobile computing and modelling counterpart, location awareness techniques, and wireless technologies have improved screen resolutions and provided technical solutions for delivering textual, 2D, and 3D information to a mobile device. After investigating small-screen representations and the related navigation content and mobile interface design from the previous cases, two suggestions are presented in this chapter in order to enhance the design for future mobile wayfinding systems: 1) pseudo-3D representations and 2) personalised and on-demand services.

Download Full-text

Assessing augmented reality in helping undergraduate students to integrate 2D and 3D representations of stereochemistry

Journal of Biocommunication ◽

10.5210/jbc.v42i1.9187 ◽

2018 ◽

Vol 42 (1) ◽

Cited By ~ 2

Author(s):

Matan Berson ◽

Derek Ng ◽

Jumi Shin ◽

Jodie Jenkinson

Keyword(s):

Organic Chemistry ◽

Augmented Reality ◽

Physical Model ◽

Undergraduate Students ◽

Mobile App ◽

Molecular Characteristics ◽

Chemistry Students ◽

Model Sets ◽

2D And 3D ◽

3D Representations

Organic chemistry students exhibit greatdifficulty grasping stereochemistry and chemicalbond hybridization concepts. The 2Drepresentations of molecules presented to themin lectures and exams serve as tools forcommunication between professional chemists.However, these students’ ability to link themolecular formulas, 2D pictures, geometricstructures, and molecular characteristics hasrepeatedly been found challenging. 2Ddepictions, physical model sets, and 3Dcomputer models are currently used inclassrooms and courses in an attempt to tacklethe problem of tying the 2D representations tothe 3D stereochemistry properties but eachtool falls short in one aspect or another. In thepresent study, we examine how an augmentedreality (AR) mobile app combining 2D and 3Drepresentations may be used to supportunderstanding of these concepts.

Download Full-text