ERUDITION ADAPTED MODEL FOR FACIAL APPEARANCE ANALYSIS AND GESTICULATION RECOGNITION

Information Technology

Image processing is the analysis and manipulation of a digital image, especially in order to improve its quality. This system is used to detect the human emotions from the input image. First an image is given as input, skin is segmented based on skin colour. The face region is then scanned by marking the edges of the face and the connected region is then cropped. From the cropped face image the eyes and the lips are separated. It draws Bezier curve for eyes and lips. Emotion of the Bezier curve is assigned as emotion of this image. The proposed approach in different applications: pain recognition and action unit detection using visual data and gestures classification using inertial measurements, demonstrating the generality of our method with respect to different input data types and basic classifiers. In terms of accuracy and computational time both with respect to user-independent approaches and to previous personalization techniques this project succeeds. This paper presents a framework for personalizing classification models which does not require labelled target data. It proposes a regression framework which exploits auxiliary annotated data to learn the relation between person-specific sample distributions and the parameters of the corresponding classifiers. Then, when considering a new target user, the classification model is computed by simply feeding the unlabelled sample distribution into the learned regression function.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.