The physical plausibility of human motions is vital to various applications in fields including but not limited to graphics, animation, robotics, vision, biomechanics, and sports science. While fully simulating human motions with physics is an extreme challenge, we hypothesize that we can treat this complexity as a black box in a data-driven manner if we focus on the ground contact, and have sufficient observations of physics and human activities in the real world. To prove our hypothesis, we present GroundLink, a unified dataset comprised of captured ground reaction force (GRF) and center of pressure (CoP) synchronized to standard kinematic motion captures. GRF and CoP of GroundLink are not simulated but captured at high temporal resolution using force platforms embedded in the ground for uncompromising measurement accuracy. This dataset contains 368 processed motion trials (∼ 1.59M recorded frames) with 19 different movements including locomotion and weight-shifting actions such as tennis swings to signify the importance of capturing physics paired with kinematics. GroundLinkNet, our benchmark neural network model trained with GroundLink, supports our hypothesis by predicting GRFs and CoPs accurately and plausibly on unseen motions from various sources. The dataset, code, and benchmark models are made public for further research on various downstream tasks leveraging the rich physics information at https://csr.bu.edu/groundlink/.
14 September 2011 in Computers and Electronics in Agriculture
In a preliminary experiment, the influence of pigment color variation on the temperature readings of the shells (EST) of live embryonated Ross 708 broiler hatching eggs was tested. Prior to set, eggshell color in L*, a*, and b* coordinates were measured at the equator of each egg. Eggshell pigmentation was found not to influence EST from zero through 18 d of incubation (DOI). In a subsequent experiment, EST, as well as air cell (ACT) and cloaca (CLT) temperature measurements were used for the calculation of absolute (GH2O) and specific (gH20) eggshell conductance values for these same types of eggs. An infrared thermometer was used to determine EST from zero to 19 DOI, ACT was measured using a transponder from 12 to 19 DOI, and CLT was determined using a transponder at 4:00 PM at 19 DOI. In the 12 to 19 DOI interval, the values for GH2O as well as for gH20 that were calculated using either EST or ACT were significantly correlated (r ≥ 0.99; P < 0.0001). A similar correlation level for both GH2O and gH20 was likewise observed in the 10:00 AM to 10:00 PM time period at 19 DOI when either EST, ACT, or CLT was used. However, in the 12 to 19 DOI interval, calculated GH2O and gH20 values based on ACT were significantly different from those based on EST. In addition, a significant difference in calculated GH2O and gH20 values resulted when ACT was used rather than when EST or CLT was used in the 10:00 AM to 10:00 PM time period at 19 DOI. In both time periods, GH2O and gH20 values calculated using ACT were significantly lower than those derived using the other 2 types of measurements. These findings suggest that although EST was not affected by shell coloration, because ACT more closely reflects embryo body temperature, it should be used to more accurately calculate the GH2O and gH20 of Ross 708 broiler hatching eggs.