SwePub
Sök i LIBRIS databas

  Utökad sökning

onr:"swepub:oai:DiVA.org:kth-290787"
 

Sökning: onr:"swepub:oai:DiVA.org:kth-290787" > Moving Fast and Slow :

LIBRIS Formathandbok  (Information om MARC21)
FältnamnIndikatorerMetadata
00003489naa a2200493 4500
001oai:DiVA.org:kth-290787
003SwePub
008210222s2021 | |||||||||||000 ||eng|
024a https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-2907872 URI
024a https://doi.org/10.1080/10447318.2021.18838832 DOI
040 a (SwePub)kth
041 a engb eng
042 9 SwePub
072 7a ref2 swepub-contenttype
072 7a art2 swepub-publicationtype
100a Kucherenko, Taras,d 1994-u KTH,Robotik, perception och lärande, RPL4 aut0 (Swepub:kth)u1igz10x
2451 0a Moving Fast and Slow :b Analysis of Representations and Post-Processing in Speech-Driven Automatic Gesture Generation
264 c 2021-02-17
264 1b Informa UK Limited,c 2021
338 a print2 rdacarrier
500 a QC 20211109
520 a This paper presents a novel framework for speech-driven gesture production, applicable to virtual agents to enhance human-computer interaction. Specifically, we extend recent deep-learning-based, data-driven methods for speech-driven gesture generation by incorporating representation learning. Our model takes speech as input and produces gestures as output, in the form of a sequence of 3D coordinates. We provide an analysis of different representations for the input (speech) and the output (motion) of the network by both objective and subjective evaluations. We also analyze the importance of smoothing of the produced motion. Our results indicated that the proposed method improved on our baseline in terms of objective measures. For example, it better captured the motion dynamics and better matched the motion-speed distribution. Moreover, we performed user studies on two different datasets. The studies confirmed that our proposed method is perceived as more natural than the baseline, although the difference in the studies was eliminated by appropriate post-processing: hip-centering and smoothing. We conclude that it is important to take both motion representation and post-processing into account when designing an automatic gesture-production method.
650 7a NATURVETENSKAPx Data- och informationsvetenskapx Människa-datorinteraktion0 (SwePub)102042 hsv//swe
650 7a NATURAL SCIENCESx Computer and Information Sciencesx Human Computer Interaction0 (SwePub)102042 hsv//eng
653 a Gesture generation
653 a representation learning
653 a neural network
653 a deep learning
653 a virtual agents
653 a non-verbal behavior
653 a Computer Science
653 a Datalogi
653 a Computer Science
653 a Datalogi
700a Hasegawa, Dai4 aut
700a Kaneko, Naoshi4 aut
700a Henter, Gustav Eje,c Assistant Professoru KTH,Robotik, perception och lärande, RPL4 aut0 (Swepub:kth)u1cj22n0
700a Kjellström, Hedvig,d 1973-u KTH,Robotik, perception och lärande, RPL4 aut0 (Swepub:kth)u1izkbhh
710a KTHb Robotik, perception och lärande, RPL4 org
773t International Journal of Human-Computer Interactiond : Informa UK Limitedg 37:14, s. 1300-1316q 37:14<1300-1316x 1044-7318x 1532-7590
856u https://doi.org/10.1080/10447318.2021.1883883y Fulltext
856u https://www.tandfonline.com/doi/pdf/10.1080/10447318.2021.1883883?needAccess=true
8564 8u https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-290787
8564 8u https://doi.org/10.1080/10447318.2021.1883883

Hitta via bibliotek

Till lärosätets databas

Sök utanför SwePub

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

 
pil uppåt Stäng

Kopiera och spara länken för att återkomma till aktuell vy