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
AES International Conference on Audio for Virtual and Augmented Reality (AVAR)
Six-degrees-of-freedom rendering of an acoustic environment can be achieved by interpolating a set of measured spatial room impulse responses (SRIRs). However, the involved measurement effort and computational expense are high. This work compares novel ways of extrapolating a single measured SRIR to a target position. The novel extrapolation techniques are based on a recently proposed subspace method that decomposes SRIRs into a direct part, comprising direct sound and salient reflections, and a residual. We evaluate extrapolations between different positions in a shoebox-shaped room in a multi-stimulus comparison test. Extrapolation using a residual SRIR and salient reflections that match the reflections at the target position is rated as perceptually most similar to the measured reference.
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
Ilkan Esiyok, Pascal Berrang, Katriel Cohn-Gordon, Robert Künnemann