Share on Facebook Tweet on Twitter Share on LinkedIn Share by email
Microsoft Research Face SDK Beta
Microsoft Research Face SDK Beta

Microsoft Research Face SDK integrates the latest face technologies from Microsoft research teams. It provides state-of-the-art algorithms to process face images, like face detection, alignment, tracking, and cartoon generation.

Details

Major features are:

Detection

Given an image, this module tries to detect all faces from it and returns a list of rectangles indicating the face positions. The detection algorithm automatically handles the illumination changes as well as different face view angles.

 

* If you find a situation where the detector cannot detect human faces under reasonable conditions, you're welcome to send us the photo (or the link to the photo).

 

Alignment

This module tries to locate the face components (eyes, brows, mouth, nose, etc) and return either component centers or component outlines.

Cartoon

This module creates a personalized cartoon portrait from a single image. User can customize the portrait by selecting different styles or applying different templates.

Tracking

This module locates the face position in real-time in a live video stream. User can use the head movement (turning left/right, etc.) to interact with a Windows Phone.

Publications

Paul Viola, Michael Jones. Rapid Object Detection using a Boosted Cascade of Simple Features. CVPR 2001.

Lin Liang, Hong Chen, Ying-Qing Xu, Heung-Yeung Shum. Example-based Caricature Generation with Exaggeration. PCCGA 2002.

R. Xiao, L. Zhu, and H. J. Zhang., Boosting Chain Learning for Object Detection, in Proceedings of International Conference on Computer Vision (ICCV03), Institute of Electrical and Electronics Engineers, Inc., October 2003.

Rong Xiao, Huaiyi Zhu, He Sun, and Xiaoou Tang, Dynamic Cascades for Face Detection, in Proceedings of International Conference on Computer Vision (ICCV07), vol. 0, pp. 1-8, IEEE Computer Society, Los Alamitos, CA, USA, 2007.

C. Zhang and P. Viola. Multiple-Instance Pruning for Learning Efficient Cascade Detectors. NIPS 2007.

Cha Zhang and Zhengyou Zhang. Winner-Take-All Multiple Category Boosting for Multi-View Face Detection. MSR-TR-2009-190.

Lin Liang, Rong Xiao, Fang Wen and Jian Sun. Face Alignment via Component based Discriminative Search. ECCV 2008.

2012 May Release

Microsoft Research Face SDK Beta for Windows Phone is available for download. Any comments or suggestions, please contact us at Microsoft Research Face SDK forum.