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
IEEE International Conference on Image Processing (IEEE ICIP)
In this paper, we present a novel rate estimation technique for rate control with frame parallelization in video encoding. The proposed rate control accounts for internal delay in the frame encoder pipeline, which is suitable for multithreaded or faster software (SW) encoders and custom hardware or ASIC (HW) encoders. It includes an accurate frame size prediction model based on a complete set of first-pass statistics data for every frame. Using better prediction models with linear and nonlinear model functions, we achieve improved accuracy, with respect to encoding quality as well as improving rate control bits-per-frame planning
Yuming Du, Robin Kips, Albert Pumarola, Sebastian Starke, Ali Thabet, Artsiom Sanakoyeu
Lisa Rivalin, Andrew Grier, Tobias Tiecke, Chi Zhou, Doris Gao, Prakriti Choudhury, John Fabian
Shreshth Tuli, Kinga Bojarczuk, Natalija Gucevska, Mark Harman, Xiao-Yu Wang, Graham Wright