Variable bit rate compression using neural network models
Номер патента: EP4278303A1
Опубликовано: 22-11-2023
Автор(ы): Amir Said, Reza Pourreza, Taco Sebastiaan COHEN, Yadong Lu, Yang Yang, Yinhao ZHU
Принадлежит: Qualcomm Inc
Опубликовано: 22-11-2023
Автор(ы): Amir Said, Reza Pourreza, Taco Sebastiaan COHEN, Yadong Lu, Yang Yang, Yinhao ZHU
Принадлежит: Qualcomm Inc
Реферат: A computer-implemented method for operating an artificial neural network (ANN) includes receiving an input. A latent representation of the input is generated. The latent representation is communicated according to a bit rate based on a learned latent scaling parameter. The latent scaling parameter is learned based on a channel index and a tradeoff parameter value that corresponds to a value that balances the bit rate and a distortion.
Variable bit rate compression using neural network models
Номер патента: US20220224926A1. Автор: Yang Yang,Yadong Lu,Amir Said,Reza Pourreza,Taco Sebastiaan COHEN,Yinhao ZHU. Владелец: Qualcomm Inc. Дата публикации: 2022-07-14.