Emotion-based image retrieval&amp;#x2014;An artificial neural network approach

Olkiewicz, Katarzyna Agnieszka; Markowska–Kaczmar, Urszula

doi:10.1109/imcsit.2010.5679722

Cited by 15 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One aspect we will have to investigate concerns the reasons for taking photographs in a certain place; the reasons can be different: something beautiful is photographed, which creates positive stimuli, but also something bad to provoke negative stimuli of social denunciation. We expect in the future to download through the API the photos taken by Flickr users and to manage them with a supervised trained neural network for the recognition of the emotional content of the images [62].…”

Section: Limitation Of Study and Topics For Further Researchesmentioning

confidence: 99%

Urban Niche Assessment: An Approach Integrating Social Media Analysis, Spatial Urban Indicators and Geo-Statistical Techniques

et al. 2020

View full text Add to dashboard Cite

Cities are human ecosystems. Understanding human ecology is important for designing and planning the built environment. The ability to respond to changes and adapt actions in a positive way helps determine the health of cities. Recently, many studies have highlighted the great potential of photographic data shared on the Flickr platform for the analysis of environmental perceptions in landscape and urban planning. Other research works used panoramic images from the Google Street View (GSV) web service to extract urban quality data. Although other researches have used social media to characterize human habitat from an emotional point of view, there is still a lack of knowledge of the correlation between environmental and physical variables of the city and visual perception, especially at a scale suitable for urban planning and design. In ecology, the environmental suitability of a territory for a given biological community is studied through species distribution models (SDM). In this work we have adopted the state of the art of SDM (the ensemble approach) to develop a methodology transferable to cities with different sizes and characteristics that uses data deriving from many sources available on a global scale: social media platform, Google internet services, shared geographical information, remote sensing and geomorphological data. The result of our application in the city of Livorno offers important information on the most significant variables for the conservation, planning and design of urban public spaces at the project scale. However, further research developments will be needed to test the model in cities of different sizes and geographic locations, integrate the model with other social media, other databases and with traditional surveys and improve the quality of indicators that can be derived from information shared on the Internet.

show abstract

Section: Limitation Of Study and Topics For Further Researchesmentioning

confidence: 99%

Urban Niche Assessment: An Approach Integrating Social Media Analysis, Spatial Urban Indicators and Geo-Statistical Techniques

et al. 2020

View full text Add to dashboard Cite

show abstract

“…When talking about emotions, it is important to mention the subjectivity inherent, since multiple emotions can appear in the same subject while looking, for example, at a picture, as well as different subjects can feel different emotions when viewing the same picture, mainly due to each subject's current emotional state and "life experiences" [21], [22]. However, the expected affective response can be considered objective, as it reflects the more-or-less unanimous response of a general audience to a given stimulus [23].…”

Section: Emotionsmentioning

confidence: 99%

Enriching IAPS and GAPED Image Datasets with Unrestrained Emotional Data

Alarcão¹,

Fonseca²

2018

International Conferences on Distributed Multimedia Systems

View full text Add to dashboard Cite

Elicitation of emotions is typically done through the presentation of emotionally salient material, like images or videos, thus requiring reliably annotated datasets. Although there are datasets with emotional information, these only describe either emotional polarities or discrete emotions. The only available dataset with both types of information restrained the participants during the study by separating a priori the images according to their polarity (positive or negative). In this paper, we describe an unrestrained study with 60 participants, where we asked them to rate the polarities and discrete emotions elicited by a set of images. The analysis of the emotional ratings made by the users revealed the most frequent correlations between the basic emotions. Furthermore, the analysis of the ratings' agreement among participants and existing datasets shows that our results are aligned with the existing ones. As a result of our study, we make available to researchers a more informative picture dataset annotated with emotional polarities and multiple emotions, as a complement to existing datasets.

show abstract

“…In [10] the authors employ assisted learning to classify diferent kinds of low-level image features like color, texture and composition with the aid of pre-classified image datasets. Qingyong Li [7], implements the classification based in direct a fuzzy representation of color in the image, whereas in [12] a neural network is trained in order to map low-level features into four groups of adjectives, five basic emotions and a positive/negative tag. In our work, given the richness of the color-emotion dataset, we chose to map emotions directly from the image pixels with further statistical processing which proved to be a robust process.…”

Section: Related Workmentioning

confidence: 99%

“…• The k th more frequent words are selected Some authors [2] [7] [12] propose the use of grading properties of color scales like HSV to perform modulation over other color atributes, for example performing a fuzzy semantics over each emotion [7]. Instead, in order to have only a discrete account of color sensation we use the CIELAB color space representation as in [15] and [17] which is closer to human perception.…”

Section: Emotion Processingmentioning

confidence: 99%

Representing pictures with emotions

Fonseca¹

2018

Preprint

View full text Add to dashboard Cite

Modern research in content-based image retrieval systems (CIBR) has become progressively more focused on the richness of human semantics. Several approaches may be used to reduced the 'semantic gap' between the high-level human experience and the low level visual features of pictures. Object ontology, among others, is one of the methods. In this paper we investigate the use of a codified emotion ontology over global color features of images to annotate the images at a high semantic level. In order to speed up the annotation process the images are sampled so that each digital image is represented by a random subset of its content. We test within controlled conditions how this random subset may represent the adequate high level emotional concept presented in the image. We monitor this information reducing process with entropy measures, showing that controlled random sampling can capture with significant relevance high level concepts for picture representation.

show abstract

Emotion-based image retrieval—An artificial neural network approach

Cited by 15 publications

References 11 publications

Urban Niche Assessment: An Approach Integrating Social Media Analysis, Spatial Urban Indicators and Geo-Statistical Techniques

Urban Niche Assessment: An Approach Integrating Social Media Analysis, Spatial Urban Indicators and Geo-Statistical Techniques

Enriching IAPS and GAPED Image Datasets with Unrestrained Emotional Data

Representing pictures with emotions

Contact Info

Product

Resources

About