Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model
Yuming Du, Robin Kips, Albert Pumarola, Sebastian Starke, Ali Thabet, Artsiom Sanakoyeu
Symposium on Applied Perception (SAP)
In this study, we analyze the performance, user preference, and sense of ownership for eight virtual grasping visualizations. Six are classified as either a tracked hand visualization or an outer hand visualization. The tracked hand visualizations are those that allow the virtual hand to enter the object being grasped, whereas the outer hand visualizations do not, thereby simulating a realistic interaction. One visualization is a compromise between the two, showing a primary virtual hand that stays outside the grasped object and a secondary hand that follows the users tracked hand into it. We use high fidelity marker-based hand tracking to control the virtual hands in real time. For each feedback technique, users repeatedly pick up a gray virtual ball, move it to a target position, and release it on the target. We found that the tracked hand visualizations result in better performance, however, the outer hand visualizations were preferred. We also found some evidence that ownership is stronger with the more realistic visualizations.
Yuming Du, Robin Kips, Albert Pumarola, Sebastian Starke, Ali Thabet, Artsiom Sanakoyeu
Bilge Acun, Benjamin Lee, Fiodar Kazhamiaka, Kiwan Maeng, Manoj Chakkaravarthy, Udit Gupta, David Brooks, Carole-Jean Wu
Harjasleen Malvai, Lefteris Kokoris-Kogias, Alberto Sonnino, Esha Ghosh, Ercan Ozturk, Kevin Lewi, Sean Lawlor