peopleDetectorACF

Detect people using aggregate channel features

Syntax

detector = peopleDetectorACF

detector = peopleDetectorACF(name)

Description

detector = peopleDetectorACF returns a pretrained upright people detector using aggregate channel features (ACF). The detector is an acfObjectDetector object, and is trained using the INRIA person data set.

detector = peopleDetectorACF(name) returns a pretrained upright people detector based on the specified model name.

Examples

Detect People Using Aggregated Channel Features

Load the upright people detector.

detector = peopleDetectorACF;

Read an image. Detect people in the image.

I = imread('visionteam1.jpg');
[bboxes,scores] = detect(detector,I);

Annotate detected people with bounding boxes and their detection scores.

I = insertObjectAnnotation(I,'rectangle',bboxes,scores);
figure
imshow(I)
title('Detected People and Detection Scores')

Input Arguments

name — ACF classification model

'inria-100x41' (default) | 'caltech-50x21'

ACF classification model, specified as 'inria-100x41' or 'caltech-50x21'. The 'inria-100x41' model was trained using the INRIA Person data set. The 'caltech-50x21' model was trained using the Caltech Pedestrian data set.

Output Arguments

detector — Trained ACF-based object detector

acfObjectDetector object

Trained ACF-based object detector, returned as an acfObjectDetector object. The detector is trained to detect upright people in an image.

References

[1] Dollar, P., R. Appel, S. Belongie, and P. Perona. "Fast Feature Pyramids for Object Detection." IEEE Transactions on Pattern Analysis and Machine Intelligence. Vol. 36, Issue 8, 2014, pp. 1532–1545.

[2] Dollar P., C. Wojek, B. Shiele, and P. Perona. "Pedestrian Detection: An Evaluation of the State of the Art." IEEE Transactions on Pattern Analysis and Machine Intelligence. Vol. 34, Issue 4, 2012, pp. 743–761.

[3] Dollar, P., C., Wojek, B. Shiele, and P. Perona. "Pedestrian Detection: A Benchmark." IEEE Conference on Computer Vision and Pattern Recognition. 2009.

Version History

Introduced in R2017a