A Color Bar Based Affective Annotation Method for Media Player

Xu, Chengzhe; Chen, Ling; Chen, Gencai

doi:10.1007/11610113_70

Cited by 6 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, in studies on affective communication in human-computer interaction (e.g., Tajadura-Jiménez and Västfjäll, 2008), the EmojiGrid could be used to enable users to repeatedly report perceived affect (e.g., by projecting a pointer-controlled beam on the grid). Such an application could also be useful for the affective annotation of multimedia (Chen et al, 2007;Soleymani et al, 2008;Runge et al, 2016;Suhaimi et al, 2018), for personalized affective multimedia retrieval (e.g., through query-by-emoji: Cappallo et al, 2019) or multimedia recommender systems (Hanjalic and Xu, 2005;Xu et al, 2008;Zhang et al, 2010;Lopatovska and Arapakis, 2011;Koelstra et al, 2012;Zhao et al, 2013), for the affective appraisal of multimedia entertainment in real-time (Fleureau et al, 2012), to give affective feedback in serious gaming applications (Anolli et al, 2010), and for affective multimedia generation (e.g., music: Kim and André, 2004). We are currently implementing the EmojiGrid in a multisensory VR environment (the Sensiks Sensory Reality Pod: www.sensiks.com) as an interface for the user to select and adjust the desired multisensory (visual, auditory, tactile, and olfactory) affective experiences.…”

Section: Discussionmentioning

confidence: 99%

An Immersive Self-Report Tool for the Affective Appraisal of 360° VR Videos

Toet

Heijn

Brouwer

et al. 2020

Front. Virtual Real.

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

An Immersive Self-Report Tool for the Affective Appraisal of 360° VR Videos

Toet

Heijn

Brouwer

et al. 2020

Front. Virtual Real.

View full text Add to dashboard Cite

“…While these existing tools require the user to successively explore the entire affective space to find the desired expression each time a response is given, the EmojiGrid provides an instantaneous overview of the affective input space. This feature may be useful for the affective annotation of multimedia ( Runge et al, 2016 ) or personalized affective video retrieval ( Xu et al, 2008 ; Lopatovska and Arapakis, 2011 ), for real-time affective evaluation of entertainment ( Fleureau et al, 2012 ) or as an affective input tool for serious gaming applications ( Anolli et al, 2010 ).…”

Section: Future Researchmentioning

confidence: 99%

EmojiGrid: A 2D Pictorial Scale for the Assessment of Food Elicited Emotions

et al. 2018

View full text Add to dashboard Cite

Research on food experience is typically challenged by the way questions are worded. We therefore developed the EmojiGrid: a graphical (language-independent) intuitive self-report tool to measure food-related valence and arousal. In a first experiment participants rated the valence and the arousing quality of 60 food images, using either the EmojiGrid or two independent visual analog scales (VAS). The valence ratings obtained with both tools strongly agree. However, the arousal ratings only agree for pleasant food items, but not for unpleasant ones. Furthermore, the results obtained with the EmojiGrid show the typical universal U-shaped relation between the mean valence and arousal that is commonly observed for a wide range of (visual, auditory, tactile, olfactory) affective stimuli, while the VAS tool yields a positive linear association between valence and arousal. We hypothesized that this disagreement reflects a lack of proper understanding of the arousal concept in the VAS condition. In a second experiment we attempted to clarify the arousal concept by asking participants to rate the valence and intensity of the taste associated with the perceived food items. After this adjustment the VAS and EmojiGrid yielded similar valence and arousal ratings (both showing the universal U-shaped relation between the valence and arousal). A comparison with the results from the first experiment showed that VAS arousal ratings strongly depended on the actual wording used, while EmojiGrid ratings were not affected by the framing of the associated question. This suggests that the EmojiGrid is largely self-explaining and intuitive. To test this hypothesis, we performed a third experiment in which participants rated food images using the EmojiGrid without an associated question, and we compared the results to those of the first two experiments. The EmojiGrid ratings obtained in all three experiments closely agree. We conclude that the EmojiGrid appears to be a valid and intuitive affective self-report tool that does not rely on written instructions and that can efficiently be used to measure food-related emotions.

show abstract

“…Since the EmojiGrid provides an instantaneous overview of the affective input space, it enables users to provide continuous affective feedback, for instance in HCI studies by moving a mousecontrolled cursor over the support of the grid. This feature may be useful for the real-time affective annotation of mulsemedia (Runge et al 2016), for personalized affective video retrieval (Lopatovska and Arapakis 2011;Xu et al 2008), for real-time affective evaluation of entertainment (Fleureau et al 2012), or as an affective input tool for serious gaming applications (Anolli et al 2010). A simplified version of the EmojiGrid has already been implemented in the Sensiks Sensory Reality Pod (www.sensiks.com) to enable the user to select a multisensory (visual, auditory, tactile, and olfactory) experience with a prespecified affective quality.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

The Relation Between Valence and Arousal in Subjective Odor Experience

Toet¹,

Eijsman

Liu

et al. 2019

Chem. Percept.

View full text Add to dashboard Cite

IntroductionThe main purpose of this study was to investigate the overall relation between the mean (at the nomothetic or group level) subjective valence and arousal ratings for odors. Although well established in other sensory modalities (e.g., visual, auditory, gustatory, tactile), this relation has not previously been investigated for odors covering a large range of the valence dimension. In addition, we evaluated the EmojiGrid (a recently introduced intuitive graphical affective self-report tool) for the affective appraisal of odors. Methods Young and healthy participants (N = 56, 32 females) used the EmojiGrid to rate the perceived valence and arousal for 40 different and randomly presented odors, ranging in valence from unpleasant to pleasant. ResultsThe overall relation between mean valence and arousal can be described by a U-shaped (quadratic) form; odors scoring near neutral on mean valence have the lowest mean arousal ratings, while odors scoring either high (pleasant) or low (unpleasant) on mean valence show higher mean arousal ratings. The results for odors that were also used in previous studies in the literature agree with their earlier reported values. Conclusion Mean arousal ratings increase with (positive or negative) emotional valence. Participants intuitively used the EmojiGrid to report their affective appraisal of odors without any verbal labels or written instructions. Implications The current findings are relevant for various applications and environments (e.g., public, retail, entertainment) where odors are used to induce desired emotional states (e.g., relaxation, arousal) and behaviors. The EmojiGrid can efficiently be applied to assess whether specific odors evoke the desired subjective affective experiences.

show abstract

A Color Bar Based Affective Annotation Method for Media Player

Cited by 6 publications

References 10 publications

An Immersive Self-Report Tool for the Affective Appraisal of 360° VR Videos

An Immersive Self-Report Tool for the Affective Appraisal of 360° VR Videos

EmojiGrid: A 2D Pictorial Scale for the Assessment of Food Elicited Emotions

The Relation Between Valence and Arousal in Subjective Odor Experience

Contact Info

Product

Resources

About