Lecture6 - VisualSimulation CAP6938 Dr.HassanForoosh Dept.ofComputerScience UCF VisualMotion Estimation DirectMethods

Info iconThis preview shows pages 1–17. Sign up to view the full content.

View Full Document Right Arrow Icon
    Visual Simulation CAP 6938 Dr. Hassan Foroosh  Dept. of Computer Science UCF © Copyright Hassan Foroosh 2002
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Visual Motion  Estimation
Background image of page 2
    Direct Methods:    Methods for motion and/or shape estimation,  which recover the unknown parameters  directly  from  measurable image quantities   at  each pixel  in the image. Minimization step:    Direct methods:    Error measure based on dense measurable image  quantities. Feature-based methods:   Error measure based on distances of destinct features.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    WHY BOTHER Visual Motion can be annoying Camera instabilities, jitter Measure it. Remove it. Visual Motion indicates dynamics in the scene Moving objects, behavior Track objects and analyze trajectories Visual Motion reveals spatial layout of the scene Motion parallax
Background image of page 4
    Video Enhancement Visual Motion can be annoying Camera instabilities, jitter Measure it. Remove it.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Temporal Information Visual Motion indicates dynamics in the  scene Moving objects, behavior Track objects and analyze trajectories
Background image of page 6
    Spatial Layout Visual Motion reveals spatial layout of the scene Motion parallax Foroosh,2002
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Classes of Techniques Feature-based methods Extract salient visual features (corners, textured areas) and track  them over multiple frames Analyze the global pattern of motion vectors of these features Sparse motion fields, but possibly robust tracking Suitable especially when image motion is large (10-s of pixels) Direct-methods Directly recover image motion from spatio temporal image  brightness variations Global motion parameters directly recovered without an  intermediate feature motion calculation Dense motion fields, but more sensitive to appearance variations Suitable for video and when image motion is small (< 10 pixels) Our Focus Today
Background image of page 8
    Brightness Constancy Equation: The Brightness  Constraint ) , ( ) , ( ) , ( ) , ( y x y x v y u x I y x J + + Or, better still, Minimize : 2 )) , ( ) , ( ( ) , ( v y u x I y x J v u E + + - = ) , ( ) , ( ) , ( ) , ( ) , ( ) , ( y x v y x I y x u y x I y x I y x J y x + + Linearizing   (assuming small  (u,v) ):
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Gradient Constraint  or the Optical Flow Constraint 2 ) ( ) , ( t y x I v I u I v u E + + = Minimizing: 0 ) ( 0 ) ( 0 = + + = + + = = t y x y t y x x I v I u I I I v I u I I dv E du E The gradient constraint – only one constraint for each pixel In general 0 , y x I I 0 + + t y x I v I u I Hence,
Background image of page 10
    Aperture Problem and  Normal Flow
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Aperture Problem and  Normal Flow
Background image of page 12
    Aperture Problem and  Normal Flow
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Aperture Problem and  Normal Flow 0 0 = = + + U I I v I u I t y x The gradient constraint: Defines a line in the (u,v) space u v I I I I u t - = Normal Flow:
Background image of page 14
    Local Patch Analysis
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Combining Local  Constraints u v 1 1 t I U I - = 2 2 t I U I - = 3 3 t I U I - = etc.
Background image of page 16
Image of page 17
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/13/2011 for the course CAP 6938 taught by Professor Staff during the Spring '08 term at University of Central Florida.

Page1 / 60

Lecture6 - VisualSimulation CAP6938 Dr.HassanForoosh Dept.ofComputerScience UCF VisualMotion Estimation DirectMethods

This preview shows document pages 1 - 17. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online