Probabilistic Bit-Rate and Rate-Distortion Cost Estimation for Video Coding
AssigneeTexas Instruments Incorporated
InventorOsman Gokhan Sezer
Life of this patent
Estimating the term…
Abstract
A method of video encoding is provided that includes computing spatial variance for video data in a block of a video sequence, estimating a first bit-rate based on the spatial variance, a transform coefficient threshold, and variance multiplicative factors empirically determined for first transform coefficients, and encoding the block based on the first bit-rate.
Claims20
A computer-implemented method of video encoding comprising:
- computing spatial variance for video data in a block of a video sequence;
- estimating a first bit-rate based on the spatial variance, a transform coefficient threshold, and variance multiplicative factors empirically determined for first transform coefficients;
- and
- encoding the block based on the first bit-rate.
The computer-implemented method of claim 2, wherein estimating a first bit-rate comprises estimating the first bit-rate as a value given by
- ϕ
- ×
- ∑
- exp
- (
- -
- 2
-
- T
- i
- σ
- s
- )
- wherein σ is a number of bits for encoding the first transform coefficients, Ti is a ratio of the transform coefficient threshold and a variance multiplicative factor of the variance multiplicative factors for the ith first transform coefficient, and σs is the spatial variance.
A digital system comprising:
- a video encoder configured to encode a block of a video sequence by computing spatial variance for video data in a block of a video sequence;
- estimating a first bit-rate based on the spatial variance, a transform coefficient threshold, and variance multiplicative factors empirically determined for first transform coefficients;
- and
- encoding the block based on the first bit-rate.
The digital system of claim 10, wherein estimating a first bit-rate comprises estimating the first bit-rate as a value given by
- ϕ
- ×
- ∑
- exp
- (
- -
- 2
-
- T
- i
- σ
- s
- )
- wherein φ is a number of bits for encoding the first transform coefficients, Ti is a ratio of the transform coefficient threshold and a variance multiplicative factor of the variance multiplicative factors for the ith first transform coefficient, and σs is the spatial variance.
The digital system of claim 10, wherein the video encoder is further configured to encode a block of a video sequence by
- estimating a second bit-rate for the block based on the spatial variance, the transform coefficient threshold, and variance multiplicative factors empirically determined for second transform coefficients, and
- wherein encoding the block comprises selecting an encoding mode based on the first bit-rate and the second-bit rate.
The digital system of claim 10, wherein the video encoder is further configured to encode a block of a video sequence by
- estimating first distortion for the block based on the spatial variance, the transform coefficient threshold, and the variance multiplicative factors empirically determined for the first transform coefficients;
- and
- estimating a first rate-distortion cost using the first bit-rate and the first distortion.
A computer readable medium comprising executable instructions to cause a digital system to perform a method of video encoding, the method comprising:
- computing spatial variance for video data in a block of a video sequence;
- estimating a first rate-distortion cost for a first encoding mode by estimating a probabilistic rate-distortion cost for each transform coefficient of first transform coefficients based on the spatial variance, a transform coefficient threshold, and variance multiplicative factors empirically determined for the first transform coefficients;
- estimating a second rate-distortion cost for a second encoding mode by estimating a probabilistic rate-distortion cost for each transform coefficient of second transform coefficients based on the spatial variance, the transform coefficient threshold, and variance multiplicative factors empirically determined for the second transform coefficients;
- and
- encoding the block using the first encoding mode when the first rate-distortion cost is less than the second rate-distortion cost.
Dependencies are inferred from the claim text.