Granted Patent B2

US 12,470,715 · App. 18/534,073 · Granted Nov 11, 2025

Inventor: Amir Said (San Diego, CA)

Assignee: QUALCOMM Incorporated

H04N19/13H04N19/124H04N19/134H04N19/136H04N19/42

Abstract

Claims (44)

1 . A method of coding media data, the method comprising:

determining a probability distribution function parameter for a data element of a data stream coded by a neural-based media compression technique, wherein the probability distribution function parameter is a function of a standard deviation of a probability distribution function of the data stream and wherein the probability distribution function parameter is based on a distribution of the data stream optimized for quantization,

determining a code vector based on the probability distribution function parameter; and

entropy coding the data element using the code vector.

2 . The method of claim 1 , further comprising:

quantizing the probability distribution function parameter prior to determining the code vector.

3 . The method of claim 1 , wherein the probability distribution function parameter is u, the standard deviation of the probability distribution function is σ, a minimum standard deviation is σ min , a maximum standard deviation is σ max , and wherein the relationship between u and σ is defined as: σ=T u→σ (u)=T λ→σ (T u→λ (u)), where T λ→σ (λ)=exp(λ ln(σ max /σ min )+ln(σ min )), where functions T are defined according to an algorithm that measures coding redundancy or defined by solving an ordinary differential equation.

4 . The method of claim 1 , further comprising:

generating the data element using the neural-based compression technique; and

quantizing the data element to create a quantized data element,

wherein entropy coding the data element using the code vector comprises entropy encoding the quantized data element using the code vector.

5 . The method of claim 4 , wherein generating the data element using the neural-based compression technique comprises:

processing an image or video picture using an image analysis neural network to generate the data element.

6 . The method of claim 5 , further comprising:

capturing the image or video picture using a camera.

7 . The method of claim 1 , wherein entropy coding the data element using the code vector comprises entropy decoding an encoded data element using the code vector to create a quantized data element, the method further comprising:

dequantizing the quantized data element to create a reconstructed data element.

8 . The method of claim 7 , further comprising:

processing the reconstructed data element using an image synthesis neural network to reconstruct an image or video picture.

9 . The method of claim 8 , further comprising:

displaying the image or the video picture.

10 . An apparatus configured to code media data, the apparatus comprising:

a memory; and

one or more processors in communication with the memory, the one or more processors configured to:

determine a probability distribution function parameter for a data element of a data stream coded by a neural-based media compression technique, wherein the probability distribution function parameter is a function of a standard deviation of a probability distribution function of the data stream and wherein the probability distribution function parameter is based on a distribution of the data stream optimized for quantization;

determine a code vector based on the probability distribution function parameter; and

entropy code the data element using the code vector.

11 . The apparatus of claim 10 , wherein the one or more processors are further configured to:

quantize the probability distribution function parameter prior to determining the code vector.

12 . The apparatus of claim 10 , wherein the probability distribution function parameter is u, the standard deviation of the probability distribution function is σ, a minimum standard deviation is σ min , a maximum standard deviation is σ max , and wherein the relationship between u and σ is defined as: σ=T u→σ (u)=T λ→σ (T u→λ (u)), where T λ→σ (λ)=exp(λ ln(σ max /σ min )+ln(σ min )), where functions T are defined according to an algorithm that measures coding redundancy or defined by solving an ordinary differential equation.

13 . The apparatus of claim 10 , wherein the one or more processors are further configured to:

generate the data element using the neural-based compression technique; and

quantize the data element to create a quantized data element,

wherein entropy coding the data element using the code vector comprises entropy encoding the quantized data element using the code vector.

14 . The apparatus of claim 13 , wherein to generate the data element using the neural-based compression technique, the one or more processors are further configured to:

process an image or video picture using an image analysis neural network to generate the data element.

15 . The apparatus of claim 14 , further comprising:

a camera configured to capture the image or video picture.

16 . The apparatus of claim 10 , wherein entropy coding the data element using the code vector comprises entropy decoding an encoded data element using the code vector to create a quantized data element, and wherein the one or more processors are further configured to:

dequantize the quantized data element to create a reconstructed data element.

17 . The apparatus of claim 16 , wherein the one or more processors are further configured to:

process the reconstructed data element using an image synthesis neural network to reconstruct an image or video picture.

18 . The apparatus of claim 17 , further comprising:

a display configured to display the image or video picture.

Assignments (1)

Continuity (3)

References Cited (102)