Global Alignment and
Structure from Motion

	Computer Vision
	CSE455, Winter 2008
	Noah Snavely

Readings

Problem: Drift

Global optimization

Global optimization

Global optimization

Global optimization

Ambiguity in global location

Solving for camera parameters

Solving for camera rotation

Solving for rotations

Solving for rotations

3D rotations

Solving for rotations and translations

Slide 15

Structure from motion

Slide 17

SfM objective function

Solving structure from motion

Photo Tourism

Photo Tourism

Photo Tourism overview

Scene reconstruction

Feature detection

Feature detection

Feature detection

	Detect features using SIFT [Lowe, IJCV 2004]

Feature matching

	Match features between each pair of images

Feature matching

Incremental structure from motion

Incremental structure from motion

Incremental structure from motion

Problem size

Photo Tourism overview

Photo Explorer

Overhead map

Prague Old Town Square

Annotations

Annotations

Yosemite

Yosemite

Slide 41

Two-view structure from motion

Notes on two-view geometry

Properties of the essential matrix

Properties of the essential matrix

The Fundamental matrix

More information