Annotated Facial Landmarks in the Wild: A large-scale, real-world database for facial landmark localization

By Martin Köstinger, Paul Wohlhart, Peter M. Roth, Horst Bischof. 2011

1,074 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

Deep High-Resolution Representation Learning for Visual Recognition (2019, methodology): “We perform the evaluation over four standard datasets: WFLW [149], AFLW [75], COFW [10], and 300 W [121].”
Turning a Blind Eye: Explicit Removal of Biases and Variation from Deep Neural Network Embeddings (2018, methodology): “We used the AFLW dataset [10] categorized by yaw, to train the pose classiﬁer mentioned in section 3.1.”
How Far are We from Solving the 2D & 3D Face Alignment Problem? (and a Dataset of 230,000 3D Facial Landmarks) (2017, methodology): “AFLW2000-3D [50] is a dataset constructed by re-annotating the first 2000 images from AFLW [21] using 68 3D landmarks in a consistent manner with the”
Face Detection without Bells and Whistles (2014, methodology): “We collected a total of 15 106 samples from the AFLW database [15] to train 5 models (components) of size 80× 80 pixels.”
Learning an animatable detailed 3D face model from in-the-wild images (2020, background): “Larsen et al. [2016] and Isola et al.”
Self-supervised learning of a facial attribute embedding from video (2018, methodology): “pared to a supervised method [26] and to using the VGG-Face descriptor. To perform the evaluation, the linear regression is trained from the given embedding to head pose labels using the AFLW dataset [24], but after leaving out the 1;000 images of the AFLW test set from [26]. As can be seen in Table”
Image and Vision Computing (2016, neutral): “The Annotated Facial Landmarks in the Wild (AFLW) [31] database consists of 25,993 images gathered from Flickr, exhibiting a large variety in appearance (e.g., pose, expression, ethnicity, age, gender) as well as general imaging and environmental conditions.”
Style Aggregated Network for Facial Landmark Detection (2018, methodology): “We further empirically verify the performance degeneration caused by the large variance of image styles.”
Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings (2017, methodology): “Due to tighter face crops, we adapt our 50-landmark CelebA network, ﬁne-tuning it ﬁrst on similarly cropped CelebA images and then on the AFLW training set.”
How We've Taught Algorithms to See Identity: Constructing Race and Gender in Image Databases for Facial Analysis (2020, background): “, Annotated Facial Landmarks in the Wild (AFLW) [85], FER-2013) and “skin type” (e.”
Facial Landmark Detection by Deep Multi-task Learning (2014, methodology): “Evaluation on AFLW [11]: Figure 9 shows that TCDCN outperforms all the state-of-the-art methods.”
From Facial Parts Responses to Face Detection: A Deep Learning Approach (2015, methodology): “(ii) For face detection training, we choose 13, 205 images from the AFLW dataset [15] to ensure a balanced out-of-plane pose distribution and 5, 771 random personfree images from the PASCAL VOC 2007 dataset.”
Unsupervised Discovery of Object Landmarks as Structural Representations (2018, methodology): “We evaluate our method on a variety of datasets, including CelebA [31] and AFLW [35] for human faces, the cat head dataset [79], a car dataset built from PASCAL 3D [69], shoe images from UT Zappos50k [74], human pose images from Human3.”
(2013) 300 Faces in-the-Wild Challenge: the first facial landmark localization challenge. In: 2013 IEEE International (2017, background): “The most well-known in-the-wild facial databases are: LFPW [2], HELEN [9], AFW [19] and AFLW [8].”