%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2017/09.11.13.40
%2 sid.inpe.br/sibgrapi/2017/09.11.13.40.19
%T Nose pose estimation in the wild and its applications on nose tracking and 3D face alignment
%D 2017
%A Zavan, Flávio Henrique de Bittencourt,
%A Silva, Luciano,
%A Bellon, Olga Regina Pereira,
%@affiliation Universidade Federal do Paraná
%@affiliation Universidade Federal do Paraná
%@affiliation Universidade Federal do Paraná
%E Torchelsen, Rafael Piccin,
%E Nascimento, Erickson Rangel do,
%E Panozzo, Daniele,
%E Liu, Zicheng,
%E Farias, Mylène,
%E Viera, Thales,
%E Sacht, Leonardo,
%E Ferreira, Nivan,
%E Comba, João Luiz Dihl,
%E Hirata, Nina,
%E Schiavon Porto, Marcelo,
%E Vital, Creto,
%E Pagot, Christian Azambuja,
%E Petronetto, Fabiano,
%E Clua, Esteban,
%E Cardeal, Flávio,
%B Conference on Graphics, Patterns and Images, 30 (SIBGRAPI)
%C Niterói, RJ, Brazil
%8 17-20 Oct. 2017
%I Sociedade Brasileira de Computação
%J Porto Alegre
%S Proceedings
%K face processing, face analysis, head pose estimation.
%X An automatic, landmark free SVM-based method for head pose estimation, solely using the nose region, in constrained and unconstrained scenarios, is presented. Using the nose region has advantages over the whole face; it is less likely to be occluded or deformed by facial expressions, and is proven to be highly discriminant in all poses from profile to frontal. The approach, SVM-NosePose, receives a nose region as and classifies it into a discrete set of poses. Estimation favorably compares against state-of-the-art works on six publicly available datasets. Three applications are derived from the proposed methodology: 1) the original inclusion of a head pose score for face quality estimation for initializing a nose tracker, leading to higher accuracy; 2) 3D face alignment in the wild using only the nose pose, enabling consistent estimates even in challenging scenarios; and 3) multipose action unit detection and intensity estimation for facial images in the wild.
%@language en
%3 wtd_sibgrapi_2017_camera_ready.pdf