Differentiable Depth for Real2Sim Calibration of Soft Body Simulations
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Differentiable Depth for Real2Sim Calibration of Soft Body Simulations. / Arnavaz, K.; Nielsen, M. Kragballe; Kry, P. G.; Macklin, M.; Erleben, K.
In: Computer Graphics Forum, Vol. 42, No. 1, 2023, p. 277-289.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Differentiable Depth for Real2Sim Calibration of Soft Body Simulations
AU - Arnavaz, K.
AU - Nielsen, M. Kragballe
AU - Kry, P. G.
AU - Macklin, M.
AU - Erleben, K.
N1 - Publisher Copyright: © 2022 The Authors. Computer Graphics Forum published by Eurographics - The European Association for Computer Graphics and John Wiley & Sons Ltd.
PY - 2023
Y1 - 2023
N2 - In this work, we present a novel approach for calibrating material model parameters for soft body simulations using real data. We use a fully differentiable pipeline, combining a differentiable soft body simulator and differentiable depth rendering, which permits fast gradient-based optimizations. Our method requires no data pre-processing, and minimal experimental set-up, as we directly minimize the L2-norm between raw LIDAR scans and rendered simulation states. In essence, we provide the first marker-free approach for calibrating a soft-body simulator to match observed real-world deformations. Our approach is inexpensive as it solely requires a consumer-level LIDAR sensor compared to acquiring a professional marker-based motion capture system. We investigate the effects of different material parameterizations and evaluate convergence for parameter optimization in both single and multi-material scenarios of varying complexity. Finally, we show that our set-up can be extended to optimize for dynamic behaviour as well.
AB - In this work, we present a novel approach for calibrating material model parameters for soft body simulations using real data. We use a fully differentiable pipeline, combining a differentiable soft body simulator and differentiable depth rendering, which permits fast gradient-based optimizations. Our method requires no data pre-processing, and minimal experimental set-up, as we directly minimize the L2-norm between raw LIDAR scans and rendered simulation states. In essence, we provide the first marker-free approach for calibrating a soft-body simulator to match observed real-world deformations. Our approach is inexpensive as it solely requires a consumer-level LIDAR sensor compared to acquiring a professional marker-based motion capture system. We investigate the effects of different material parameterizations and evaluate convergence for parameter optimization in both single and multi-material scenarios of varying complexity. Finally, we show that our set-up can be extended to optimize for dynamic behaviour as well.
KW - animation
KW - methods and applications
KW - physically based animation
KW - ray tracing
KW - rendering
KW - robotics
U2 - 10.1111/cgf.14720
DO - 10.1111/cgf.14720
M3 - Journal article
AN - SCOPUS:85143054771
VL - 42
SP - 277
EP - 289
JO - Computer Graphics Forum (Print)
JF - Computer Graphics Forum (Print)
SN - 0167-7055
IS - 1
ER -
ID: 339158200