TY - GEN
T1 - Personalized emotion-aware video streaming for the elderly
AU - Dong, Yi
AU - Hu, Han
AU - Wen, Yonggang
AU - Yu, Han
AU - Miao, Chunyan
N1 - Publisher Copyright:
© Springer International Publishing AG, part of Springer Nature 2018.
PY - 2018
Y1 - 2018
N2 - We consider the problem of video therapy services for the elderly based on their current emotional status. Given long hours watching TV in the elder population, most of the existing TV services are not geared for them. The elderly cannot tolerate complexity and negativity due to decline in cognitive abilities. In addition, the program is not adapted to the user’s current emotional status. As a result, existing TV services can not achieve optimal performance across a broad set of user types and context. To provide content tailored to individual needs, and interests of the elderly, caregivers have to select an appropriate program manually. However, this can not scale well due to shortage of caregivers and high monetary cost. We present the personalized emotion-aware video streaming system, a redesign of conventional TV system to provide appropriate program flexibly, efficiently and responsively. Our proposed architecture adds video affective profiling, real-time emotion detection and Markov decision process based video program generation to the streaming service to this end. We present a complete implementation of our design. Trace-driven simulation has shown the effectiveness of our system.
AB - We consider the problem of video therapy services for the elderly based on their current emotional status. Given long hours watching TV in the elder population, most of the existing TV services are not geared for them. The elderly cannot tolerate complexity and negativity due to decline in cognitive abilities. In addition, the program is not adapted to the user’s current emotional status. As a result, existing TV services can not achieve optimal performance across a broad set of user types and context. To provide content tailored to individual needs, and interests of the elderly, caregivers have to select an appropriate program manually. However, this can not scale well due to shortage of caregivers and high monetary cost. We present the personalized emotion-aware video streaming system, a redesign of conventional TV system to provide appropriate program flexibly, efficiently and responsively. Our proposed architecture adds video affective profiling, real-time emotion detection and Markov decision process based video program generation to the streaming service to this end. We present a complete implementation of our design. Trace-driven simulation has shown the effectiveness of our system.
UR - http://www.scopus.com/inward/record.url?scp=85050560754&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-91485-5_28
DO - 10.1007/978-3-319-91485-5_28
M3 - Conference contribution
AN - SCOPUS:85050560754
SN - 9783319914848
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 372
EP - 382
BT - Social Computing and Social Media. Technologies and Analytics - 10th International Conference, SCSM 2018, Held as Part of HCI International 2018, Proceedings
A2 - Meiselwitz, Gabriele
PB - Springer Verlag
T2 - 10th International Conference on Social Computing and Social Media, SCSM 2018 Held as Part of HCI International 2018
Y2 - 15 July 2018 through 20 July 2018
ER -