Constrained local neural fields for robust facial landmark detection in the wild (bibtex)
by Baltrušaitis, Tadas, Robinson, Peter and Morency, Louis-Philippe
Abstract:
Facial feature detection algorithms have seen great progress over the recent years. However, they still struggle in poor lighting conditions and in the presence of extreme pose or occlusions. We present the Constrained Local Neural Field model for facial landmark detection. Our model includes two main novelties. First, we introduce a probabilistic patch expert (landmark detector) that can learn non-linear and spatial relationships between the input pixels and the probability of a landmark being aligned. Secondly, our model is optimised using a novel Non-uniform Regularised Landmark Mean-Shift optimisation technique, which takes into account the reliabilities of each patch expert. We demonstrate the benefit of our approach on a number of publicly available datasets over other state-of-the-art approaches when performing landmark detection in unseen lighting conditions and in the wild.
Reference:
Constrained local neural fields for robust facial landmark detection in the wild (Baltrušaitis, Tadas, Robinson, Peter and Morency, Louis-Philippe), In Computer Vision Workshops (ICCVW), 2013 IEEE International Conference on, IEEE, 2013.
Bibtex Entry:
@inproceedings{baltrusaitis_constrained_2013,
	address = {Sydney, Australia},
	title = {Constrained local neural fields for robust facial landmark detection in the wild},
	url = {http://ict.usc.edu/pubs/Constrained%20local%20neural%20fields%20for%20robust%20facial%20landmark%20detection%20in%20the%20wild.pdf},
	abstract = {Facial feature detection algorithms have seen great progress over the recent years. However, they still struggle in poor lighting conditions and in the presence of extreme pose or occlusions. We present the Constrained Local Neural Field model for facial landmark detection. Our model includes two main novelties. First, we introduce a probabilistic patch expert (landmark detector) that can learn non-linear and spatial relationships between the input pixels and the probability of a landmark being aligned. Secondly, our model is optimised using a novel Non-uniform Regularised Landmark Mean-Shift optimisation technique, which takes into account the reliabilities of each patch expert. We demonstrate the benefit of our approach on a number of publicly available datasets over other state-of-the-art approaches when performing landmark detection in unseen lighting conditions and in the wild.},
	booktitle = {Computer {Vision} {Workshops} ({ICCVW}), 2013 {IEEE} {International} {Conference} on},
	publisher = {IEEE},
	author = {Baltrušaitis, Tadas and Robinson, Peter and Morency, Louis-Philippe},
	month = dec,
	year = {2013},
	keywords = {Virtual Humans, UARC},
	pages = {354--361}
}
Powered by bibtexbrowser