Understanding images of groups of people

By Andrew C. Gallagher, Tsuhan Chen. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009

400 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/21d9d0deed16f0ad62a4865e9acf0686f4f15492

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

Age Synthesis and Estimation via Faces: A Survey (2010, methodology): “The Web-collected database reported in [10] contains 28,231 faces (in 5,080 images) from the Flickr.com image search engineusingthree-groupsearchessuchas“wedding+bride+ groom+portrait” “group shot” or “group photo” or “group portrait” “family portrait.””
Age and gender classification using convolutional neural networks (2015, methodology): “…manual filtering, as is typically the case on media webpages (e.g., images from the LFW collection [25]) or social websites (the Group Photos set [14]), viewing conditions in these images are highly unconstrained, reflecting many of the real-world challenges of faces appearing in Internet…”
From Facial Expression Recognition to Interpersonal Relation Prediction (2016, background): “Human Interaction and Group Behavior Analysis There exists a number of studies that analyze human interaction and group behavior from images and videos (Ding and Yilmaz 2010, 2011; Fathi et al. 2012; Ramanathan et al. 2013; Ricci et al. 2015; Weng et al. 2009; Gallagher and Chen 2009).”
Gender Classification by Deep Learning on Millions of Weakly Labelled Images (2016, methodology): “Additionally, we wished to test our performance on the GROUPS dataset [10], a set of 28,231 faces annotated with age, gender and location information from a collection of 5,080 group photos.”
Where's Waldo: Matching people in images of crowds (2011, background): “We also note that prior authors have explored color models for matching people [7, 11, 13], co-occurrence cues [6], and other contextual cues [8, 13] in other settings.”
Urban tribes: Analyzing group photos from a social perspective (2012, background): “This includes an approach for capturing the structure of the group to provide better interpretation of the image [7] or the extension of this work to recognize social relationships between pairs of people, such as family relationships [16].”
AFIF4: Deep Gender Classification based on AdaBoost-based Fusion of Isolated Facial Features and Foggy Faces (2017, methodology): “In order to detect multiple faces in a single image (as in the case of the Groups dataset), we repeat the following 2-step procedure: 1) we apply the CDSM to detect a face; 2) we mask out the detected face region by a single-color rectangle and repeat step 1 again on the same image until no more fac”
AgeDB: The First Manually Collected, In-the-Wild Age Database (2017, background): “Gallagher [10] 2009 28,231 5,080 7 age groups No Yes”
DAGER: Deep Age, Gender and Emotion Recognition Using Convolutional Neural Network (2017, methodology): “We followed the setup in [5] where 3 , 500 images are used for training and 1 , 050 images are used for testing.”
Describing Clothing by Semantic Attributes (2012, methodology): “We adopt the publicly available gender-from-face implementation of [33], which outputs gender probabilities by projecting aligned faces in the Fisherface space.”
Beyond frontal faces: Improving Person Recognition using multiple cues (2015, background): “In [12], the authors investigated the different factors that are related to the positions of people in a group image.”
Understanding and Comparing Deep Neural Networks for Age and Gender Classification (2017, background): “8% to reportedly 91% [7] on the recent and challenging Adience benchmark [8], mirroring the overall progress on other available benchmarks such as the Images of Groups data set [12], the LFW data set [18] or the Ghallagher Collection Person data set [11].”
Age and Gender Estimation of Unfiltered Faces (2014, methodology): “Mid: The Gallagher collection [5], provides images with an intentional bias towards groups of people, typically facing the camera and posing for their shots.”
Learning ordinal discriminative features for age estimation (2012, methodology): “Besides comparing with related feature selection algorithms, we also compare with the method in [10], which utilizes contextual features for age range estimation.”
Age , Gender and Race Estimation from Unconstrained Face Images (2014, methodology): “To the best of our knowledge, Gallagher and Chen [8] were the first to study demographic (age group and gender) estimation from unconstrained face images.”
VIP: Finding important people in images (2015, background): “We found that when subjects were posed an absolute question “Please mark the important people in this image,” they found the task difﬁ-cult.”
Local Deep Neural Networks for gender recognition (2016, methodology): “(2007)) and the so-called Gallagher’s database (Gallagher and Chen (2009)), where the images were taken in unconstrained conditions.”
ChaLearn Looking at People and Faces of the World: Face AnalysisWorkshop and Challenge 2016 (2016, neutral): “GROUPS [19] 28,231 28,231 0 - 66+ Age group No No -”