Affdex SDK

McDuff, Daniel; Mahmoud, Abdelrahman; Mavadati, Mohammad; Amr, May; Turcot, Jay; Kaliouby, Rana el

doi:10.1145/2851581.2890247

Cited by 233 publications

(41 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[14]), and workload (Driving Activity Load Index [38]). We furthermore analyzed the driver's facial expressions regarding displayed emotions using the Affdex SDK [30] and assessed the propriety of the used matching algorithm, i.e., whether or not the recommended character matched with the preferred character. At the end of the experiment, participants answered questions on the experienced characters in a semi-structured interview and assessed the perceived personalities using a semantic differential rating.…”

Section: Methodsmentioning

confidence: 99%

“…We evaluate the output of the Affdex facial expression detection system [30] for a 10-second time frame after each interaction. This duration was chosen as literature states emotional responses to auditory stimuli are processed against preexisting expectations within less than 10 seconds after Table 3: Emotion detection values for a 10-second time frame after experiencing the characters.…”

Section: Emotion Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

At Your Service

Braun

Mainz

Chadowitz

et al. 2019

Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems

138

View full text Add to dashboard Cite

DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the "Taverne" license above, please follow below link for the End User Agreement:

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Emotion Recognitionmentioning

confidence: 99%

At Your Service

Braun

Mainz

Chadowitz

et al. 2019

Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems

138

View full text Add to dashboard Cite

show abstract

“…Affdex performs automatic facial coding in four steps: face and facial landmark detection, face feature extraction, facial action, and emotion expression modeling based on the EMFACS emotional facial action coding system (Ekman and Friesen, 1978; Friesen and Ekman, 1983; McDuff et al, 2016). Although no data has been published yet specifically comparing the performance of the software on adults vs. children, FACS coding is generally the same for adults and for children and has been used with children as young as 2 years (e.g., Camras et al, 2006; LoBue and Thrasher, 2015; also see Ekman and Rosenberg, 1997).…”

Section: Methodsmentioning

confidence: 99%

“…Children’s affect data were collected using Affdex whenever a face was detected with the front-facing camera on the Samsung Galaxy S4 device (McDuff et al, 2016). Affdex is capable of measuring 15 expressions, which are used to calculate the likelihood that the detected face is displaying each of nine different affective states.…”

Section: Methodsmentioning

confidence: 99%

Flat vs. Expressive Storytelling: Young Children’s Learning and Retention of a Social Robot’s Narrative

Westlund

Jeong

Park

et al. 2017

Front. Hum. Neurosci.

View full text Add to dashboard Cite

Prior research with preschool children has established that dialogic or active book reading is an effective method for expanding young children’s vocabulary. In this exploratory study, we asked whether similar benefits are observed when a robot engages in dialogic reading with preschoolers. Given the established effectiveness of active reading, we also asked whether this effectiveness was critically dependent on the expressive characteristics of the robot. For approximately half the children, the robot’s active reading was expressive; the robot’s voice included a wide range of intonation and emotion (Expressive). For the remaining children, the robot read and conversed with a flat voice, which sounded similar to a classic text-to-speech engine and had little dynamic range (Flat). The robot’s movements were kept constant across conditions. We performed a verification study using Amazon Mechanical Turk (AMT) to confirm that the Expressive robot was viewed as significantly more expressive, more emotional, and less passive than the Flat robot. We invited 45 preschoolers with an average age of 5 years who were either English Language Learners (ELL), bilingual, or native English speakers to engage in the reading task with the robot. The robot narrated a story from a picture book, using active reading techniques and including a set of target vocabulary words in the narration. Children were post-tested on the vocabulary words and were also asked to retell the story to a puppet. A subset of 34 children performed a second story retelling 4–6 weeks later. Children reported liking and learning from the robot a similar amount in the Expressive and Flat conditions. However, as compared to children in the Flat condition, children in the Expressive condition were more concentrated and engaged as indexed by their facial expressions; they emulated the robot’s story more in their story retells; and they told longer stories during their delayed retelling. Furthermore, children who responded to the robot’s active reading questions were more likely to correctly identify the target vocabulary words in the Expressive condition than in the Flat condition. Taken together, these results suggest that children may benefit more from the expressive robot than from the flat robot.

show abstract

“…Facial expressions recognition algorithms infer emotions using both geometric (position of nose, eyes, mouth), and appearance (pixels values) features of the face. These emotions are used by some tools such as FaceReader (Loijens and Krips, 2019), Microsoft Emotion API (Microsoft.com, n.d.), Affdex SDK (McDuff et al, 2016) and Google Emotion API (Google.com, n.d.) to recognize emotions by providing numerical values 5 associated to them. Some tools show high accuracy in detecting basic emotions, but only on specific datasets (e.g.…”

Section: Emotion Recognition and Learning Analyticsmentioning

confidence: 99%

Emodash: A dashboard supporting retrospective awareness of emotions in online learning

Ez-Zaouia

Tabard

Lavoué

2020

International Journal of Human-Computer Studies

View full text Add to dashboard Cite

We present Emodash, an interactive dashboard supporting tutors' retrospective awareness of learners' emotions in a video-conferencing learning environment.Socio-affective relationships play an important role in learning processes and learning outcomes, but they are harder to develop in online-learning. This can be explained by a lack of emotion awareness due to the asynchronous interactions, technical challenges, and tutors' focus on properly conducting the learning activity and gearing towards pedagogical outcomes. We conducted an eight-week long field study with five professional tutors on how they used Emodash while writing feedback to learners after language learning sessions. We found that Emodash led tutors who were already sensitive to learners' emotions to incorporate more affective elements in their reports, suggesting a stronger awareness of learners' emotions. Tutors also wrote more formative and less summative feedback. Furthermore, our results suggest that glanceable visualizations of learners' emotions may be preferred and sufficient to foster tutors' awareness of learners' emotions. Finally, the dashboard led tutors to reflect on the way they conduct their lessons, using learners' positive emotions as a proxy evaluation of their teaching.

show abstract

Affdex SDK

Cited by 233 publications

References 9 publications

At Your Service

At Your Service

Flat vs. Expressive Storytelling: Young Children’s Learning and Retention of a Social Robot’s Narrative

Emodash: A dashboard supporting retrospective awareness of emotions in online learning

Contact Info

Product

Resources

About