Affective image classification using features inspired by psychology and art theory

By Jana Machajdik, Allan Hanbury. 2010

903 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition, Social Psychology

How researchers see this paper

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

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

In Social Psychology

Citations in the corpus

Exploring Principles-of-Art Features For Image Emotion Recognition (2014, methodology): “However, there are several disadvantages using them to model image emotions: (1) Weak link to emotions [1, 26].”
Automatic Understanding of Image and Video Advertisements (2017, background): “[57, 58, 6, 42, 30, 45] study the emotions shown or perceived in images, but for generic images, rather than ones purposefully created to convey an emotion.”
Predicting Pre-click Quality for Native Advertisements (2016, methodology): “Textures GLCM Properties [22] 4 Entropy, Energy, Contrast, and Homogeneity of the Gray-Level Co-”
Do points, levels and leaderboards harm intrinsic motivation?: an empirical analysis of common gamification elements (2013, methodology): “The image annotation task consisted of 15 abstract paintings that were taken from Machajdik and Hanbury’s study on affective image classification [15].”
Towards understanding the effects of individual gamification elements on intrinsic motivation and performance (2017, methodology): “The image annotation task consisted of 15 abstract paintings Please cite this article in press as: Mekler, E. D., et al., Towards unders motivation and performance, Computers in Human Behavior (2015), http that were taken from Machajdik and Hanbury (2010)'s study on affective image classification.”
Interpretable aesthetic features for affective image classification (2013, methodology): “We adopt eight emotional categories, amusement, anger, awe, contentment, disgust, excitement, fear and sad [1] on two public data sets: (1) ARTphoto: a set of 806 artistic photos and (2) ABSTRACT: a set of 228 abstract paintings, which are the same as those in [1].”
Building a Large Scale Dataset for Image Emotion Recognition: The Fine Print and The Benchmark (2016, methodology): “For example, based on art theory (Itten and Van Haagen 1973; Valdez and Mehrabian 1994), Machajdik and Hanbury (Machajdik and Hanbury 2010) deﬁned eight different kinds of pixel level features (e.g. color, texture and composition), which have been empirically proved to be related to emotional reacti”
DeepSentiBank: Visual Sentiment Concept Classification with Deep Convolutional Neural Networks (2014, background): “There is also work trying to model “non-conventional” concepts or properties of the images, such as image aesthetic and quality [16, 22], memorability [12], interestingness [12], and affection/emotions [21, 35, 13, 21, 35, 37].”
Streetscore -- Predicting the Perceived Safety of One Million Streetscapes (2014, methodology): “Early research [4, 16] in this area was based on features extracted using rules from photography theory and psychology.”
Understanding and Predicting Image Memorability at a Large Scale (2015, methodology): “To create a varied dataset, we sampled images from a number of existing datasets such as MIR Flickr [11], AVA dataset [27], affective images dataset [25] (consisting of Art and Abstract datasets), image saliency datasets (MIT1003 [14] and NUSEF [28]), SUN [34], image popularity dataset [16], Abnorma”
The Interestingness of Images (2013, methodology): “Machadjik and Hanbury [17] extracted emotion scores from raw pixels.”
Large-scale visual sentiment ontology and detectors using adjective noun pairs (2013, methodology): “Further, as shown earlier the application of special features such as aesthetic features used in [21] and face expression features offers interesting potential for further improvement.”