Smoothing techniques for 3D animated human pose estimation data
(2024) In Master's Theses in Mathematical Sciences FMAM05 20241Mathematics (Faculty of Engineering)
- Abstract
- Virtual 3D reconstructions of live sport events are on the horizon and to produce
a high quality experience for viewers it is important that the movements of the 3D
models look natural. Today, state of the art pose estimators produce data that con-tains noise, resulting in jittery animations with pose errors. The goals for this thesis were to evaluate the performance of a Long Short-Term Memory (LSTM) Neural Network and classical filters in their ability to reduce such noise, and if they can be used to improve the viewer experience. This was done by comparing the Mean Per Joint Positional Error (MPJPE) and Absolute Acceleration Error (AAE) for
artificially added noise to motion capture data and comparing the results. A qual-
itative... (More) - Virtual 3D reconstructions of live sport events are on the horizon and to produce
a high quality experience for viewers it is important that the movements of the 3D
models look natural. Today, state of the art pose estimators produce data that con-tains noise, resulting in jittery animations with pose errors. The goals for this thesis were to evaluate the performance of a Long Short-Term Memory (LSTM) Neural Network and classical filters in their ability to reduce such noise, and if they can be used to improve the viewer experience. This was done by comparing the Mean Per Joint Positional Error (MPJPE) and Absolute Acceleration Error (AAE) for
artificially added noise to motion capture data and comparing the results. A qual-
itative survey was also constructed to evaluate how the different methods affected
the human perception of the animations when focusing on different aspects. We
concluded that our methods were (to varying degrees) able to reduce the amount
of noise introduced to the motion data. The qualitative study also indicated that the Savitzky-Golay smoothing algorithm improved the participants perception of the
animations. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9173997
- author
- Lundgren, Oliver LU and Bergöö, August
- supervisor
- organization
- course
- FMAM05 20241
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Pose Estimation, 3D, Animations, Data Smoothing and Filtering, Sports, Machine Learning, Neural Networs, Spiideo
- publication/series
- Master's Theses in Mathematical Sciences
- report number
- LUTFMA-3547-2024
- ISSN
- 1404-6342
- other publication id
- 2024:E48
- language
- English
- id
- 9173997
- date added to LUP
- 2024-09-30 14:25:32
- date last changed
- 2024-09-30 14:25:32
@misc{9173997, abstract = {{Virtual 3D reconstructions of live sport events are on the horizon and to produce a high quality experience for viewers it is important that the movements of the 3D models look natural. Today, state of the art pose estimators produce data that con-tains noise, resulting in jittery animations with pose errors. The goals for this thesis were to evaluate the performance of a Long Short-Term Memory (LSTM) Neural Network and classical filters in their ability to reduce such noise, and if they can be used to improve the viewer experience. This was done by comparing the Mean Per Joint Positional Error (MPJPE) and Absolute Acceleration Error (AAE) for artificially added noise to motion capture data and comparing the results. A qual- itative survey was also constructed to evaluate how the different methods affected the human perception of the animations when focusing on different aspects. We concluded that our methods were (to varying degrees) able to reduce the amount of noise introduced to the motion data. The qualitative study also indicated that the Savitzky-Golay smoothing algorithm improved the participants perception of the animations.}}, author = {{Lundgren, Oliver and Bergöö, August}}, issn = {{1404-6342}}, language = {{eng}}, note = {{Student Paper}}, series = {{Master's Theses in Mathematical Sciences}}, title = {{Smoothing techniques for 3D animated human pose estimation data}}, year = {{2024}}, }