reu week iii

16
REU WEEK III Malcolm Collins-Sibley Mentor: Shervin Ardeshir

Upload: dennis

Post on 22-Feb-2016

57 views

Category:

Documents


0 download

DESCRIPTION

REU Week III. Malcolm Collins-Sibley Mentor: Shervin Ardeshir. Project. Cross-View Image Registration and Semantic Segmentation - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: REU Week III

REU WEEK IIIMalcolm Collins-SibleyMentor: Shervin Ardeshir

Page 2: REU Week III

PROJECT• Cross-View Image Registration and

Semantic Segmentation• The goal is to use information from map

and satellite images, and project them on the screen which the user is observing, in a way that the user can see semantic segments overlaid on the scene.

Page 3: REU Week III

PROJECT• Output Mockup

Page 4: REU Week III

COMPLETED WORK• Readings:

• “Geometric Image Parsing in Man-Made Environments”• Olga Barinova et al

• “Recovering Surface Layout from an Image”• Derek Hoiem et al

• “Recovering Occlusion Boundaries from a Single Image”• Derek Hoeim et al

• “Entropy Rate Superpixel Segmentation”• MY Liu et al

Page 5: REU Week III

COMPLETED WORK• Geometric Image Parsing Code

Page 6: REU Week III

COMPLETED WORK• Geometric Image Parsing Code

Page 7: REU Week III

COMPLETED WORK• Super-pixel Segmentation

With 8 super-pixels

Page 8: REU Week III

COMPLETED WORK• Super-pixel Segmentation

With 20 super-pixels

Page 9: REU Week III

COMPLETED WORK• Building Projection

Page 10: REU Week III

COMPLETED WORK• Building Projection

Page 11: REU Week III

CURRENT WORK• Within the Building Projection code:

• Building occlusion and self-occlusion• Works when a building occludes another,

but not when a building is occluding itself

Page 12: REU Week III

CURRENT WORK• Occlusion Handling

Before After

Page 13: REU Week III

CURRENT WORK• Occlusion HandlingBefore After

Page 14: REU Week III

CURRENT WORK• Occlusion HandlingBefore After

Page 15: REU Week III

CURRENT WORK• Occlusion HandlingBefore After

Page 16: REU Week III

THE NEXT STEP• Understanding the occlusion handling code

• Making sure it is handling self-occlusions accurately

• Understanding the format of the output data in the line segments/horizon code

• Running the line segmentation code for all of the images in our dataset and saving all of the output variables in a structure

• Extracting the super pixels from images in the dataset and saving it in a structure

• Computing their pairwise similarities of the super pixels in terms of color and texture