Kids Pose Project

Kids Pose Project

We are looking to discover and characterize differences in how children and adults make natural and prompted body movements when using whole-body interaction systems such as the Microsoft Kinect. Little is known about how different child motion is from adult motion, and whether the differences will actually impact interaction. We ran the first explicit study of the perception of child motion compared to analogous adult motion and we found that people can generally identify whether a motion was performed by a child or an adult significantly above chance and with even higher accuracy levels for dynamic motions, such as walking and running; our findings was published in the ACM Transactions on Applied Perception [PDF]. Findings from our perception study establish that there are perceivable differences between children’s and adults’ motions. To quantify these differences, we initially focused on analyzing walking and running motions from our Kinder-Gator dataset, a dataset of children and adults performing motions forward-facing the Kinect. To analyze these motions, we relied on temporal (dependent on time) and spatial (dependent on distance) features from the gait literature. Our findings showed that children differ from adults only in terms of temporal features; children generally complete walking and running motions in less time and with higher energy compared to adults. Our findings from this analysis was published in the International Conference on Human-Computer Interaction [PDF]. These findings show that there exist features that can quantify the differences between children’s and adults’ motions. Our future work on this project will focus on identifying features that quantify the differences between how children and adults move in general (i.e., independent of the type of motion being performed). We are working on the project with Eakta Jain at UF.

Check out the full project website!

Pose Publications

Journal Articles

1. Jain, E., Anthony, L., Aloba, A., Castonguay, A., Cuba, I., Shaw, A., and Woodward, J. 2016. Is the motion of a child perceivably different from the motion of an adult? ACM Transactions on Applied Perception, Volume 13, Issue 4, Article 22, July 2016. [PDF]

Refereed Conference Papers

2. Aloba, A., Luc, A., Woodward,J., Dong, Y., Zhang, R., Jain, E., and Anthony, L. 2019. Quantifying Differences between Child and Adult Motion Based on Gait Features. In International Conference on Human-Computer Interaction , Orlando, FL, USA, July 29-31, Pages 385-402. Springer, Cham. [PDF]
3. Aloba, A., Flores, G., Woodward, J., Shaw, A., Castonguay, A., Cuba, I., Dong, Y., Jain, E.and Anthony, L. 2018. Kinder-Gator: The UF Kinect Database of Child and Adult Motion. EUROGRAPHICS Proceedings, Delft, Netherlands, April 16-20, 2018, 4 pages. [PDF]
4. Dong, Y., Paryani, S., Rana, N., Aloba, A., Anthony, L., Jain, E. 2017. Adult2Child: dynamic scaling laws to create child-like motion. MIG’17 , November 8–10, 2017, Barcelona, Spain, pages 1-10. [PDF]

The Team

Aishat Aloba
Dr. Lisa Anthony
Yuzhu Dong (Member from Jain Lab)
Dr. Eakta Jain
Alex Shaw
Julia Woodward
Amanda Castonguay (Former Member)
Isabella Cuba (Former Member from Jain Lab)

Funding

This work is partially supported by National Science Foundation Grant Awards #IIS-1552598. Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect these agencies’ views.

Blogposts