Granted Patent B2

US 12,022,077 · App. 18/230,318 · Granted Jun 25, 2024

Inventors: Chri Besenbruch (London, GB); Ciro Cursio (London, GB); Christopher Finlay (London, GB); Vira Koshkina (London, GB); Alexander Lytchier (London, GB); Jan Xu (London, GB); Arsalan Zafar (London, GB)

Assignee: DEEP RENDER LTD.

H04N19/126G06N3/045G06N3/084G06T3/4046G06T9/002G06V10/774H04N19/13

Abstract

Claims (30)

1. 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 an invertible neural network (INN), 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; and

(vii) the second computer system using an inverse of the INN to produce an output image from the quantized latent, wherein the output image is an approximation of the input image.

2. The method of claim 1 , wherein the INN comprises one or more normalising flow layers including one or more of: additive coupling layers; multiplicative coupling layers; affine coupling layers; invertible 1×1 convolution layers.

3. The method of claim 2 , wherein the one or more normalising flow layers use a continuous normalising flow.

4. The method of claim 2 , wherein the one or more normalising flow layers use a discrete normalising flow.

5. The method of claim 1 , wherein one or more weights of the INN are compressed with a normalising flow and transmitted along within the bitstream.

6. The method of claim 1 , wherein encoding the input image using the INN includes using one or more univariate or multivariate Padé activation units.

7. 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 y latent representation;

(iii) quantizing the y latent representation using the first computer system to produce a quantized y latent;

(iv) encoding the quantized y latent using an invertible neural network (INN), using the first computer system, to produce a z latent representation;

(v) quantizing the z latent representation using the first computer system to produce a quantized z latent;

(vi) entropy encoding the quantized z latent into a second bitstream, using the first computer system;

(vii) the first computer system processing the quantized z latent using an inverse of the INN to obtain probability distribution parameters of each element of the quantized y latent, wherein the probability distribution of the quantized y latent is assumed to be represented by a probability distribution of each element of the quantized y latent;

(viii) entropy encoding the quantized y latent, using the obtained probability distribution parameters of each element of the quantized y latent, into a first bitstream, using the first computer system;

(ix) transmitting the first bitstream and the second bitstream to a second computer system;

(x) the second computer system entropy decoding the second bitstream to produce the quantized z latent;

(xi) the second computer system processing the quantized z latent using a trained neural network identical to the inverse of the INN to obtain the probability distribution parameters of each element of the quantized y latent;

(xii) the second computer system using the obtained probability distribution parameters of each element of the quantized y latent, together with the first bitstream, to obtain the quantized y latent; and

(xiii) the second computer system using a second trained neural network to produce an output image from the quantized y latent, wherein the output image is an approximation of the input image.

8. The method claim 7 , wherein the INN and the inverse of the INN comprise a hyperprior network comprising a normalising flow component.

9. The method of claim 8 , wherein the normalising flow component comprises a plurality of factor-out blocks and an output of each factor-out block is processed by the hyperprior network.

10. The method of claim 8 , wherein one or more weights of the INN are compressed with a normalising flow and transmitted along within the bitstreams.

Assignments (1)

Priority Claims (12)

Continuity (6)

Cited By (3)

US 12,315,229 US 12,327,382 US 12,647,611