Reading 3: iCanDraw? – Using Sketch Recognition and Corrective Feedback to Assist a User in Drawing Human Faces

Citation

Dixon, Daniel, Manoj Prasad, and Tracy Hammond. "iCanDraw: using sketch recognition and corrective feedback to assist a user in drawing human faces."Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. ACM, 2010.

PDF

Summary

System designed to help people draw human faces. Human face in an image is first pre-processed using facial recognition algorithms to extract major feature points of the face. These facial points are then used to provide suggestions and similarity scores to users. Huge emphasis is on teaching style and how to guide user to draw better images. The teaching style adopted in this paper is step-by-step instruction coupled with timely corrective feedback.

Discussion

The reference images is pre-processed to extract 40 facial features. These 40 point features are used to set a template which is used to match and evaluate the user's drawn sketch. A similarity measures is established by considering a sliding window of 4 points in the user drawn sketch and calculating the distance between the closest 4 points in the set template. The user is given feedback based on the similarity score.