Patent · US Active

Retargeting skeleton motion sequences through cycle consistency adversarial training of a motion synthesis neural network with a forward kinematics layer

US10546408B2 · kind B2 · utility

3Cited by
7References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMar 20, 2018
Grant dateJan 28, 2020
Priority date
Expiry dateMar 28, 2038

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06T2213/12
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

This disclosure relates to methods, non-transitory computer readable media, and systems that use a motion synthesis neural network with a forward kinematics layer to generate a motion sequence for a target skeleton based on an initial motion sequence for an initial skeleton. In certain embodiments, the methods, non-transitory computer readable media, and systems use a motion synthesis neural network comprising an encoder recurrent neural network, a decoder recurrent neural network, and a forward kinematics layer to retarget motion sequences. To train the motion synthesis neural network to retarget such motion sequences, in some implementations, the disclosed methods, non-transitory computer readable media, and systems modify parameters of the motion synthesis neural network based on one or both of an adversarial loss and a cycle consistency loss.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.