TY - GEN
T1 - A discriminative parts based model approach for fiducial points free and shape constrained head pose normalisation in the wild
AU - Dhall, Abhinav
AU - Sikka, Karan
AU - Littlewort, Gwen
AU - Goecke, Roland
AU - Bartlett, Marian
PY - 2014
Y1 - 2014
N2 - This paper proposes a method for parts-based view-invariant head pose normalisation, which works well even in difficult real-world conditions. Handling pose is a classical problem in facial analysis. Recently, parts-based models have shown promising performance for facial landmark points detection 'in the wild'. Leveraging on the success of these models, the proposed data-driven regression framework computes a constrained normalised virtual frontal head pose. The response maps of a discriminatively trained part detector are used as texture information. These sparse texture maps are projected from non-frontal to frontal pose using block-wise structured regression. Finally, a facial kinematic shape constraint is achieved by applying a shape model. The advantages of the proposed approach are: a) no explicit dependence on the outputs of a facial parts detector and, thus, avoiding any error propagation owing to their failure; (b) the application of a shape prior on the reconstructed frontal maps provides an anatomically constrained facial shape; and c) modelling head pose as a mixture-of-parts model allows the framework to work without any prior pose information. Experiments are performed on the Multi-PIE and the 'in the wild' SFEW databases. The results demonstrate the effectiveness of the proposed method.
AB - This paper proposes a method for parts-based view-invariant head pose normalisation, which works well even in difficult real-world conditions. Handling pose is a classical problem in facial analysis. Recently, parts-based models have shown promising performance for facial landmark points detection 'in the wild'. Leveraging on the success of these models, the proposed data-driven regression framework computes a constrained normalised virtual frontal head pose. The response maps of a discriminatively trained part detector are used as texture information. These sparse texture maps are projected from non-frontal to frontal pose using block-wise structured regression. Finally, a facial kinematic shape constraint is achieved by applying a shape model. The advantages of the proposed approach are: a) no explicit dependence on the outputs of a facial parts detector and, thus, avoiding any error propagation owing to their failure; (b) the application of a shape prior on the reconstructed frontal maps provides an anatomically constrained facial shape; and c) modelling head pose as a mixture-of-parts model allows the framework to work without any prior pose information. Experiments are performed on the Multi-PIE and the 'in the wild' SFEW databases. The results demonstrate the effectiveness of the proposed method.
UR - http://www.scopus.com/inward/record.url?scp=84904646370&partnerID=8YFLogxK
U2 - 10.1109/WACV.2014.6835991
DO - 10.1109/WACV.2014.6835991
M3 - Conference contribution
SN - 9781479949854
T3 - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
SP - 1028
EP - 1035
BT - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
PB - IEEE Computer Society
T2 - 2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
Y2 - 24 March 2014 through 26 March 2014
ER -