Computer Vison at ETHZ - ETH Zurich | Flashcards & Summaries

Select your language

Suggested languages for you:
Log In Start studying!

Lernmaterialien für Computer Vison an der ETHZ - ETH Zurich

Greife auf kostenlose Karteikarten, Zusammenfassungen, Übungsaufgaben und Altklausuren für deinen Computer Vison Kurs an der ETHZ - ETH Zurich zu.

TESTE DEIN WISSEN

What is the vanishing point?

Lösung anzeigen
TESTE DEIN WISSEN

Its where parallel lines meet in a picture.

Lösung ausblenden
TESTE DEIN WISSEN

What are features in an image that convey depth information?

Lösung anzeigen
TESTE DEIN WISSEN

shadows, texture, focus/blurs, parallel/converging lines, lighting shading, occlusion, size/scale

Lösung ausblenden
TESTE DEIN WISSEN

What are the extrinsic parameters of camera calibration?

Lösung anzeigen
TESTE DEIN WISSEN

Its where the camera 3D coordinate system is with respect to the world coordinate system.

Lösung ausblenden
TESTE DEIN WISSEN

Why is the epipolar constraint useful?

Lösung anzeigen
TESTE DEIN WISSEN

Because it reduces the correspondence problem (finding two corresponding pixels) to a 1D search along an epipolar line.

Lösung ausblenden
TESTE DEIN WISSEN

Beyond the hard constraint of epipolar geometry what are  "soft" constraints to help identify corresponding points?

Lösung anzeigen
TESTE DEIN WISSEN
  1. Similarity (points look similar)
  2. Uniqueness (no more than one match)
  3. Ordering (same ordering of points in the images)
  4. Disparity gradient is limited (depth doesn't change too quickly)
Lösung ausblenden
TESTE DEIN WISSEN

How do you find correspondences with the similarity constraint?

Lösung anzeigen
TESTE DEIN WISSEN

With an image patch, you iterate over the image and calculate their SSD (sum of squared differences) and where there is a global minimum is where the match is.

Lösung ausblenden
TESTE DEIN WISSEN

Whats disparity?

Lösung anzeigen
TESTE DEIN WISSEN

The disparity is the distance a point has moved with respect to another picture.

Lösung ausblenden
TESTE DEIN WISSEN

What are the intrinsic parameters of camera calibration?

Lösung anzeigen
TESTE DEIN WISSEN

It's the parameters that define the projection from the 3D coordinates in the camera coordinate system to the 2D image plane.

Lösung ausblenden
TESTE DEIN WISSEN

What is a rigid body?

Lösung anzeigen
TESTE DEIN WISSEN

It's a collection of points whose relative position can't change.

Lösung ausblenden
TESTE DEIN WISSEN

How many degrees of freedom does a rigid body transformation have in 3D?

Lösung anzeigen
TESTE DEIN WISSEN

6 because 3 are for fixing the position of one point and 3 are for rotating around each axis.

Lösung ausblenden
TESTE DEIN WISSEN

What is A x A for any A and where x is the cross product?

Lösung anzeigen
TESTE DEIN WISSEN

0

Lösung ausblenden
TESTE DEIN WISSEN

What is stereo geometry?

Lösung anzeigen
TESTE DEIN WISSEN

It is that we have multiple views of the same thing

Lösung ausblenden
  • 100871 Karteikarten
  • 1794 Studierende
  • 88 Lernmaterialien

Beispielhafte Karteikarten für deinen Computer Vison Kurs an der ETHZ - ETH Zurich - von Kommilitonen auf StudySmarter erstellt!

Q:

What is the vanishing point?

A:

Its where parallel lines meet in a picture.

Q:

What are features in an image that convey depth information?

A:

shadows, texture, focus/blurs, parallel/converging lines, lighting shading, occlusion, size/scale

Q:

What are the extrinsic parameters of camera calibration?

A:

Its where the camera 3D coordinate system is with respect to the world coordinate system.

Q:

Why is the epipolar constraint useful?

A:

Because it reduces the correspondence problem (finding two corresponding pixels) to a 1D search along an epipolar line.

Q:

Beyond the hard constraint of epipolar geometry what are  "soft" constraints to help identify corresponding points?

A:
  1. Similarity (points look similar)
  2. Uniqueness (no more than one match)
  3. Ordering (same ordering of points in the images)
  4. Disparity gradient is limited (depth doesn't change too quickly)
Mehr Karteikarten anzeigen
Q:

How do you find correspondences with the similarity constraint?

A:

With an image patch, you iterate over the image and calculate their SSD (sum of squared differences) and where there is a global minimum is where the match is.

Q:

Whats disparity?

A:

The disparity is the distance a point has moved with respect to another picture.

Q:

What are the intrinsic parameters of camera calibration?

A:

It's the parameters that define the projection from the 3D coordinates in the camera coordinate system to the 2D image plane.

Q:

What is a rigid body?

A:

It's a collection of points whose relative position can't change.

Q:

How many degrees of freedom does a rigid body transformation have in 3D?

A:

6 because 3 are for fixing the position of one point and 3 are for rotating around each axis.

Q:

What is A x A for any A and where x is the cross product?

A:

0

Q:

What is stereo geometry?

A:

It is that we have multiple views of the same thing

Computer Vison

Erstelle und finde Lernmaterialien auf StudySmarter.

Greife kostenlos auf tausende geteilte Karteikarten, Zusammenfassungen, Altklausuren und mehr zu.

Jetzt loslegen

Das sind die beliebtesten StudySmarter Kurse für deinen Studiengang Computer Vison an der ETHZ - ETH Zurich

Für deinen Studiengang Computer Vison an der ETHZ - ETH Zurich gibt es bereits viele Kurse, die von deinen Kommilitonen auf StudySmarter erstellt wurden. Karteikarten, Zusammenfassungen, Altklausuren, Übungsaufgaben und mehr warten auf dich!

Das sind die beliebtesten Computer Vison Kurse im gesamten StudySmarter Universum

Computer Vision

RWTH Aachen

Zum Kurs
Computer Vision

Universität Stuttgart

Zum Kurs
Computer Vision

Leibniz Universität Hannover

Zum Kurs
Computer Vision

RWTH Aachen

Zum Kurs

Die all-in-one Lernapp für Studierende

Greife auf Millionen geteilter Lernmaterialien der StudySmarter Community zu
Kostenlos anmelden Computer Vison
Erstelle Karteikarten und Zusammenfassungen mit den StudySmarter Tools
Kostenlos loslegen Computer Vison