Learning to Align from Scratch

By Gary B. Huang, Marwan A. Mattar, Honglak Lee, Eric Learned-Miller. Neural Information Processing Systems, 2012

358 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/76fdf16bcc2cb260b9e6b2880c8fe128533bc2c6

How researchers see this paper

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

Citations in the corpus

Do Convnets Learn Correspondence? (2014, background): “Deep congealing [7] builds on this idea by replacing hand-engineered features with unsupervised feature learning from multiple resolutions.”
Adapting Models to Signal Degradation using Distillation (2016, methodology): “On this dataset a baseline would be to remove the distortion by alignment methods such as congealing [24], or use a model that estimates deformations during learning, such as spatial transformer networks [25].”
Labeled Faces in the Wild : Updates and New Reporting Procedures (2014, methodology): “The second [5] was used to produce the “deep funneled” version of LFW.”
Hierarchical Subquery Evaluation for Active Learning on a Graph (2014, unknown): “ 0.680 0.832 0.903 0.896 0.840 0.896 FlickrMat [28] 1000 50 10 PCA BoW 0.172 0.131 0.125 0.242 0.261 0.244 0.249 0.259 Coil20 [24] 1440 20 20 PCA 0.558 0.392 0.456 0.713 0.729 0.757 0.756 0.760 LFW10 [15] 1456 50 10 PCA BoW 0.310 0.261 0.247 0.352 0.421 0.419 0.410 0.422 UIUCSport [21] 1579 50 8 PCA”
When Face Recognition Meets with Deep Learning: An Evaluation of Convolutional Neural Networks for Face Recognition (2015, methodology): “The images we used are aligned by deep funneling [11].”
Local Deep Neural Networks for gender recognition (2016, methodology): “The Labeled Faces in the Wild (LFW) (Huang et al. (2007)) is composed by 13233 face images (10256 male and 2977 female) from 5749 celebrities collected from the web.”
Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings (2017, methodology): “[27] learn joint alignment of faces using deep features, and Jaiswal et al.”
DeepFace: Closing the Gap to Human-Level Performance in Face Verification (2014, background): “3D model of the face [28, 32, 14], (2) searching for similar fiducial-points configurations from an external dataset to infer from [4], and (3) unsupervised methods that find a similarity transformation for the pixels [17, 15].”
Age and Gender Estimation of Unfiltered Faces (2014, methodology): “More recently, deep learning was used for image alignment in [64].”
Pedestrian Alignment Network for Large-scale Person Re-Identification (2017, methodology): “[48] propose an unsupervised method called funneled image to align faces according to the distribution of other images and improve this method with convolutional RBM descriptor later [49].”
Labeled Faces in the Wild: A Survey (2016, methodology): “An improved version of congealing was developed [44], and was used to produce another version of LFW, known as the deep-funneled version.”
Past, Present, and Future of Face Recognition: A Review (2020, methodology): “Three types of align d images are pr posed: (1) t e funneled images [27], (2) LFW-A used an unpublished method for alignmen , and (3) deep funneled images [28].”
Augmenting CRFs with Boltzmann Machine Shape Priors for Image Labeling (2013, unknown): “, [17, 11, 12]).”
Effective face frontalization in unconstrained images (2014, background): “…that pool information over large image regions, thereby accounting for possible misalignments due to pose changes (e.g., [20, 25, 33, 36]), by improving 2D face alignment accuracy [14, 15, 35], or by using massive face collections to learn pose-robust representations [13, 26, 27, 41].”
Learning to Disentangle Factors of Variation with Manifold Interaction (2014, methodology): “Most previous work focused on building (Lowe, 1999) or learning (Kavukcuoglu et al., 2009; Ranzato et al., 2007; Lee et al., 2011; Le et al., 2011; Huang et al., 2012b;a; Sohn & Lee, 2012) invariant features that are unaffected Proceedings of the 31 st International Conference on Machine Learning,…”