Real-time facial feature detection using conditional regression forests

By Matthias Dantone, Juergen Gall, Gabriele Fanelli, Luc Van Gool. 2012

433 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, background): “[6] show that the incorporation of additional information can enhance the dependency between output variables and latent variables in a random forest framework.”
Robust Face Landmark Estimation under Occlusion (2013, methodology): “For LFW, instead of using a fixed training/test set as in the other two, the evaluation procedure proposed in [12] consists of a ten-fold cross validation using each time 1,500 training images and the rest for testing.”
Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings (2017, background): “Other approaches directly regress the landmark coordinates [59, 14, 10, 52].”
Alternating Regression Forests for Object Detection and Pose Estimation (2013, background): “This mode of operation has led to state-of-the-art in many computer vision applications, such as, facial fiducial detection [4, 6] or pose estimation of heads [7, 8] and humans [10].”
Unsupervised Discovery of Object Landmarks as Structural Representations (2018, methodology): “For human faces, there are active appearance models [10, 38, 11], template-based methods [42, 83], regression-based methods [61, 12, 7, 48], and more recent methods based on deep neural networks [55, 77, 81, 82, 75, 70, 68, 33, 71].”
COUNT Forest: CO-Voting Uncertain Number of Targets Using Random Forest for Crowd Density Estimation (2015, background): “The idea of probabilistic voting was also exploited in object tracking [18], action recognition [28], facial feature detection [12], and human pose estimation [13].”
Deep Convolutional Network Cascade for Facial Point Detection (2013, methodology): “Regressors are often used as the predictor, based on local patches close to the facial point [9] [26], or the whole image region [5] [25].”
A Comprehensive Analysis of Deep Regression (2018, neutral): “In computer vision, regression techniques span a large ensemble of applicative scenarios such as: head-pose estimation [1], [2], facial landmark detection [3], [4], human pose estimation [5], [6], age estimation [7], [8], or image registration [9], [10].”
Survey on RGB, 3D, Thermal, and Multimodal Approaches for Facial Expression Recognition: History, Trends, and Affect-related Applications (2016, methodology): “The real-time method of [80] uses conditional regression forests (CRF) over a dense grid, extracting intensity features and Gabor wavelets at each cell.”
Pose for Action - Action for Pose (2016, methodology): “We then obtain the kernel matrix that we use for the multi-class classification as follows K(vi, vj) = exp ( − 1 L L∑ t=1 Dt(h t i, h t j) µt ) (10) where L is the number of descriptor types and µt is the mean of the distance matrix Dt.”
One-Shot Person Re-identification with a Consumer Depth Camera (2014, methodology): “Once the face has been detected, a realtime method to extract the 2D location of 10 fiducials points is applied [11].”
Fisher Vector Faces in the Wild (2013, neutral): “[8] conditional random forests, and Zhu et al.”
Learning Deep Representation for Face Alignment with Auxiliary Attributes (2014, background): “We believe that facial landmark detection is not a standalone problem, but its estimation can be influenced by a number of heterogeneous and subtly correlated factors.”
Facial Landmark Detection: A Literature Survey (2018, methodology): “Different from Dantone et al. (2012) which merges the prediction from different pose-dependent models, in testing, it predicts the attributes first and then performs attribute dependent landmark location estimation.”
Human Pose Estimation Using Body Parts Dependent Joint Regressors (2013, methodology): “Random forests [5] or in general decision forests [6] have been used for many classification or regression tasks, for instance, labeling body parts in depth images [29], predicting the joint positions from depth data [15], or localizing facial feature points [8].”
Video Object Segmentation without Temporal Information (2017, background): “[11] builds a conditional random forest to estimate face landmarks whose classifier is dependent on the pose of head.”
One millisecond face alignment with an ensemble of regression trees (2014, background): “Our work builds on the large amount of research over the last decade that has resulted in significant progress for face alignment [9, 4, 13, 7, 15, 1, 16, 18, 3, 6, 19].”
Conditional regression forests for human pose estimation (2012, methodology): “[6] have used a conditional regression forests model for facial feature detection.”
A Survey on Human Motion Analysis from Depth Data (2013, methodology): “In recent years, there has been a shift towards methods based on independent local feature detectors [74, 75, 76, 77].”