公开(公告)号：US12081759B2

公开(公告)日：2024-09-03

申请号：US18230240

申请日：2023-08-04

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/00 ,  G06V10/774 ,  H04N19/13

CPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/002 ,  G06V10/774 ,  H04N19/13

摘要： There is disclosed a computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; (ii) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products are disclosed.

公开(公告)号：US11881003B2

公开(公告)日：2024-01-23

申请号：US18099444

申请日：2023-01-20

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： G06T9/00 ,  G06N3/084 ,  G06N3/044 ,  G06N3/045 ,  G06N3/047 ,  G06N3/088 ,  H04N19/59 ,  G06N20/10

CPC分类号： G06T9/002 ,  G06N3/084

摘要： A computer-implemented method of training an image generative network fθ for a set of training images, in which an output image {circumflex over (x)} is generated from an input image x of the set of training images non-losslessly, and in which a proxy network is trained for a gradient intractable perceptual metric that evaluates a quality of an output image {circumflex over (x)} given an input image x, the method of training using a plurality of scales for input images from the set of training images. In an embodiment, a blindspot network bα is trained which generates an output image {tilde over (x)} from an input image x. Related computer systems, computer program products and computer-implemented methods of training are disclosed.

公开(公告)号：US11843777B2

公开(公告)日：2023-12-12

申请号：US18105338

申请日：2023-02-03

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Aleksandar Cherganski ,  Christopher Finlay ,  Alexander Lytchier ,  Jonathan Rayner ,  Tom Ryder ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N11/02 ,  H04N19/13 ,  H04N19/124 ,  H04N19/42 ,  G06V10/422

CPC分类号： H04N19/13 ,  G06V10/422 ,  H04N19/124 ,  H04N19/42

摘要： Lossy or lossless compression and transmission, comprising the steps of: (i) receiving an input image; (ii) encoding it to produce a y latent representation; (iii) encoding the y latent representation to produce a z hyperlatent representation; (iv) quantizing the z hyperlatent representation to produce a quantized z hyperlatent representation; (v) entropy encoding the quantized z hyperlatent representation into a first bitstream, (vi) processing the quantized z hyperlatent representation to obtain a location entropy parameter μy, an entropy scale parameter σy, and a context matrix Ay of the y latent representation; (vii) processing the y latent representation, the location entropy parameter μy and the context matrix Ay, to obtain quantized latent residuals; (viii) entropy encoding the quantized latent residuals into a second bitstream; and (ix) transmitting the bitstreams.

公开(公告)号：US11677948B2

公开(公告)日：2023-06-13

申请号：US17740716

申请日：2022-05-10

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/08 ,  H04N19/13 ,  G06V10/774 ,  G06N3/04 ,  G06N3/084

CPC分类号： H04N19/126 ,  G06N3/0454 ,  G06N3/084 ,  G06V10/774 ,  H04N19/13

摘要： There is disclosed a computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; ({umlaut over (υ)}) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products are disclosed.

公开(公告)号：US12075053B2

公开(公告)日：2024-08-27

申请号：US18230376

申请日：2023-08-04

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/00 ,  G06V10/774 ,  H04N19/13

CPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/002 ,  G06V10/774 ,  H04N19/13

摘要： There is disclosed a computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; (ii) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products are disclosed.

公开(公告)号：US12028525B2

公开(公告)日：2024-07-02

申请号：US18230312

申请日：2023-08-04

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/00 ,  G06V10/774 ,  H04N19/13

CPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/002 ,  G06V10/774 ,  H04N19/13

摘要： There is disclosed a computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; (ii) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products are disclosed.

公开(公告)号：US12015776B2

公开(公告)日：2024-06-18

申请号：US18230314

申请日：2023-08-04

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/00 ,  G06V10/774 ,  H04N19/13

CPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/002 ,  G06V10/774 ,  H04N19/13

摘要： A computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; (ii) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products.

公开(公告)号：US11893762B2

公开(公告)日：2024-02-06

申请号：US18055621

申请日：2022-11-15

申请人： DEEP RENDER LTD

发明人： Thomas Ryder ,  Alexander Lytchier ,  Vira Koshkina ,  Christian Besenbruch ,  Arsalan Zafar

IPC分类号： G06N3/08 ,  G06T9/00 ,  G06V10/25 ,  G06T3/40 ,  G06N3/045

CPC分类号： G06T9/002 ,  G06N3/045 ,  G06N3/08 ,  G06T3/4046 ,  G06V10/25

摘要： A method for lossy image or video encoding, transmission and decoding, the method comprising the steps of: receiving an input image at a first computer system; encoding the input image using a first trained neural network to produce a latent representation; identifying one or more regions of the input image associated with high visual sensitivity; encoding the one or more regions of the input image associated with high visual sensitivity using a second trained neural network to produce one or more region latent representations; performing a quantization process on the latent representation and the one or more region latent representations; transmitting the result of the quantization process to a second computer system; decoding the result of the quantization process to produce an output image, wherein the output image is an approximation of the input image.

公开(公告)号：US12095994B2

公开(公告)日：2024-09-17

申请号：US18230255

申请日：2023-08-04

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/00 ,  G06V10/774 ,  H04N19/13

CPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/002 ,  G06V10/774 ,  H04N19/13

摘要： There is disclosed a computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; (ii) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products are disclosed.

公开(公告)号：US12022077B2

公开(公告)日：2024-06-25

申请号：US18230318

申请日：2023-08-04

申请人： DEEP RENDER LTD

发明人： Chri Besenbruch ,  Ciro Cursio ,  Christopher Finlay ,  Vira Koshkina ,  Alexander Lytchier ,  Jan Xu ,  Arsalan Zafar

IPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/00 ,  G06V10/774 ,  H04N19/13

CPC分类号： H04N19/126 ,  G06N3/045 ,  G06N3/084 ,  G06T3/4046 ,  G06T9/002 ,  G06V10/774 ,  H04N19/13

摘要： There is disclosed a computer-implemented method for lossy image or video compression, transmission and decoding, the method including the steps of: (i) receiving an input image at a first computer system; (ii) encoding the input image using a first trained neural network, using the first computer system, to produce a latent representation; (iii) quantizing the latent representation using the first computer system to produce a quantized latent; (iv) entropy encoding the quantized latent into a bitstream, using the first computer system; (v) transmitting the bitstream to a second computer system; (vi) the second computer system entropy decoding the bitstream to produce the quantized latent; (vii) the second computer system using a second trained neural network to produce an output image from the quantized latent, wherein the output image is an approximation of the input image. Related computer-implemented methods, systems, computer-implemented training methods and computer program products are disclosed.