Improve viola jones face detection

Hello, I am trying to detect multiple faces with Matlab built-in viola jones face detection.But in some cases it gives me false detection. Is that possible to train my own classifier and then integrate into the Matlab classification model itself in oder to reduce the false detections?
thank you.

1 Kommentar

LAXMI NARAYAN SONI
LAXMI NARAYAN SONI am 1 Apr. 2017
You Can Apply Threshold on boundary Box. it removes the small part detection. You can Also Apply Margin in viola Jones Directory In This Case.

Melden Sie sich an, um zu kommentieren.

Antworten (1)

Walter Roberson
Walter Roberson am 27 Feb. 2016

1 Stimme

Yes, the VJ face detection is a particular trained detector. You can train your own detector within exactly the same setup; see http://www.mathworks.com/help/vision/ug/train-a-cascade-object-detector.html

4 Kommentare

amir
amir am 28 Feb. 2016
Thank You. but can I combine the output of my trained detector with the one already exist in Matlab in order to get better result?
Dima Lisin
Dima Lisin am 29 Feb. 2016
You can run both detectors, and only keep the detections that they agree on.
amir
amir am 29 Feb. 2016
Bearbeitet: amir am 29 Feb. 2016
I have used the "FrontalFaceCart" and "ProfileFace" in parallel and in some cases I got overlapping boxes. How can I merge them into one? And any idea why it's not detecting the second person in the attached photo?
David Quincoces
David Quincoces am 22 Jan. 2021
do you still need help?

Melden Sie sich an, um zu kommentieren.

Kategorien

Mehr zu Image Processing and Computer Vision finden Sie in Hilfe-Center und File Exchange

Gefragt:

am 27 Feb. 2016

Kommentiert:

am 22 Jan. 2021

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by