Method of detecting facial attributes

A processing system to perform image analysis processing, comprising:

• a face detection component to analyze an image to detect a face in the image;
• a facial landmark detection component to analyze the facial image to detect facial landmarks;
• an aligning and normalizing component to align and normalize the facial image based at least in part on the detected facial landmarks to produce a normalized facial image;
• an extraction component to extract a plurality of local features from selected local regions of the normalized facial image;
• a prediction component to predict the facial attribute in each selected local region by inputting each selected local feature into one of a plurality of weak classifier components, each weak classifier component having a multi-layer perceptron (MLP) structure;
• and
• an aggregation component to output data from each weak classifier component to generate an indication that the facial attribute is detected in the facial image.

A processing system to perform image analysis processing, comprising:

• a camera to capture an image;
• a face detection component to analyze the image to detect a face in the image;
• a facial landmark detection component to analyze the facial image to detect facial landmarks;
• an aligning and normalizing component to align and normalize the facial image based at least in part on the detected facial landmarks to produce a normalized facial image;
• an extraction component to extract a plurality of local features from selected local regions of the normalized facial image;
• a prediction component to predict the facial attribute in each selected local region by inputting each selected local feature into one of a plurality of weak classifier components, each weak classifier component having a multi-layer perceptron (MLP) structure;
• an aggregation component to output data from each weak classifier component to generate an indication that the facial attribute is detected in the facial image;
• and
• a display to show the image and the indication.

Non-transitory computer-readable instructions arranged, when executed, to implement a method or realize an apparatus having:

• a facial landmark detection component to analyze the facial image to detect facial landmarks;
• an aligning and normalizing component to align and normalize the facial image based at least in part on the detected facial landmarks to produce a normalized facial image;
• an extraction component to extract a plurality of local features from selected local regions of the normalized facial image;
• a prediction component to predict the facial attribute in each selected local region by inputting each selected local feature into one of a plurality of weak classifier components, each weak classifier component having a multi-layer perceptron (MLP) structure;
• and
• an aggregation component to output data from each weak classifier component to generate an indication that the facial attribute is detected in the facial image.