Waldo Face Detection
Date published:
Description
We used the computer vision OpenCV framework to accomplish this face detector. We trained a Support Vector Model to recognize faces in Where’s Waldo images. The image features that are used by the SVM are: color histograms, HOG descriptors and SIFT descriptors. The algorithm uses a sliding window of varying sizes to search for faces in the Where’s Waldo image. If the SVM detects a face, the sliding window bounding box is marked.
Role
Programming: Geert Beuneker.
Details
Framework: OpenCV
Language: C++
Link: Source Code