Top Banner
College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences Near Near Real-Time Real-Time Cutting Cutting Paul F. Neumann
20

College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

Jan 03, 2016

Download

Documents

Shavonne Ball
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Near Near

Real-TimeReal-Time

CuttingCutting

Paul F. Neumann

Page 2: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Virtual Reality Surgical Simulators

• Simulate the functionality of surgical instruments such as blades and scissors

• A general 3D cutting algorithm is a one of challenging problems.

• Simulators must maintain an interactive frame rate.

Page 3: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Previous Cutting Algorithms

1988 Tearing (Terzopoulos and Fleischer)1992 Particle Systems (Szeliski and Tonnesen)1992 Radial Projection on FEM (Pieper et al.)1995 2D FEM Template (Song and Reddy)1997 3D FEM with Bilinear Cutting Plane (Mazura and Seifert)1997 Boolean Operations (Delp et al.)1998 Hybrid Elastic Model (Colin et al.)

My Goal: To develop an interactive cutting algorithm on a mass-spring system with a polygonal surface.

Page 4: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Mass-Spring System

•Very popular PBM platform

•Vertices as mass points

•Edges as vector springs

•Dynamic system which permits

insertions and deletions

•Distributes mass appropriately

•Conserves surface area

Page 5: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Vector Springs

• Invented by Alan Millman at EVL

• Maintain their orientation and length

• Easy to subdivide

Page 6: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Algorithm Overview

1) Samples blade’s path.

2) Reconstructs the path with a series of parallelograms.

3) Intersects and subdivides springs and triangles.

4) Recomputes mass and spring stiffness coefficients.

Page 7: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Reconstructing Cutting Path

• Discretely samples path.

• Drop samples if roughly co-planar.

• Fit parallelogram through two selected samples by

averaging orientation and adding offset.

• Parallelograms lag behind current position.

Page 8: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Intersection

Page 9: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Subdivision

New interior springs must compute their rest directionthrough vector addition of their neighbors.

Intersection SpringSubdivision

TriangleSubdivision

FurtherSubdivision

Page 10: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Mass Distribution:Localized Approximation

• Mass proportional to surface area at rest.

• New vertices and their neighbors have their mass values recomputed after subdivision.

• Spring rest direction vectors outline the undeformed triangle area.

Page 11: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Cutting Example

Page 12: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Cutting Example

Page 13: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Cutting Example

Page 14: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Cutting Example

Page 15: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Cutting Example

Page 16: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Cutting Example

Page 17: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Variations

Suction Cutter Tearing

Scissors

Page 18: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Video Tape

Page 19: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

Discussion

• Geometry dependence.

• Lag in response time.

• Rounds to nearest vertex.

• Collision Detection

• Small parallelograms within a

triangle aren’t processed.

Page 20: College of Health and Human Development Sciences, University of Illinois at Chicago Dept. of Ophthalmology and Visual Sciences NearReal-TimeCutting Paul.

College of Health and Human Development Sciences, University of Illinois at Chicago

Dept. of Ophthalmology and Visual Sciences

More Information

A more detailed paper is included on

your Application cdrom.

Web Site:

www.bvis.uic.edu/paul/

CAL Demonstration right after session