Back to results
US 2014/0233636 A1Published application
Adaptive coding of a prediction error in hybrid video coding
OwnerMatthias Narroschke, Hans-Georg Musmanninventors
Life of this patent
Estimating the term…
Abstract
The present invention relates to a method for coding a video signal using hybrid coding, comprising: reducing temporal redundancy by block based motion compensated prediction in order to establish a prediction error signal, deciding whether to transform the prediction error signal into the frequency domain, or to maintain the prediction error signal in the spatial domain for encoding.
Claims28
A method for coding a video signal using hybrid coding, the method comprising:
- reducing temporal redundancy by block based motion compensated prediction in order to establish a prediction error signal,
- transforming the prediction error signal into a transformed signal in the frequency domain,
- scalar quantizing the transformed signal into a quantized transformed signal,
- scalar quantizing the prediction error signal into a quantized prediction error signal,
- deciding whether to encode the quantized transformed signal, or encode the quantized prediction error signal in the spatial domain,
- encoding the quantized transformed signal in the frequency domain in response to deciding to encode the quantized transformed signal, and
- encoding the quantized prediction error signal in the spatial domain in response to deciding to encode the quantized prediction error signal.
4 more elements
A method for decoding a video signal using hybrid decoding, comprising:
- receiving coded video data that include coded frequency domain data and coded spatial domain data,
- decoding the coded frequency domain data to obtain decoded quantized frequency domain data and decoding the coded spatial domain data to obtain decoded quantized spatial domain data;
- inverse scalar quantizing the decoded quantized frequency domain data to obtain frequency domain data;
- inverse scalar quantizing the decoded quantized spatial domain data to obtain spatial domain data;
- performing an inverse transform of the frequency domain data from the frequency into the spatial domain, and
- skipping an inverse transform of the spatial domain data.
3 more elements
A coder for coding a video signal using hybrid coding, comprising:
- means for reducing temporal redundancy by block based motion compensated prediction and establishing a prediction error signal,
- a transform block configured to transform the prediction error signal into a transformed signal in the frequency domain,
- a first scalar quantizer configured to scalar quantize the transformed signal into a quantized transformed signal,
- a second scalar quantizer configured to scalar quantize the prediction error signal into a quantized prediction error signal,
- adaptive control means for deciding whether to encode the quantized transformed signal, or to encode the quantized prediction error signal in the spatial domain, and
- an encoder configured to encode the quantized transformed signal in the frequency domain in response to deciding to encode the quantized transformed signal, the encoder being configured to encode the quantized prediction error signal in the spatial domain in response to deciding to encode the quantized prediction error signal in the spatial domain.
3 more elements
A decoder for decoding a signal using hybrid decoding, comprising:
- receiving means configured to receive a coded signal encoded by use of hybrid coding and including coded frequency domain data and coded spatial domain data;
- adaptive control means for adaptively determining whether received coded data represents prediction error signal of the coded video signal in the spatial domain or in the frequency domain;
- and
- inverse quantization means for performing an inverse quantization of the received coded data;
- wherein the inverse quantization means includes a scalar inverse quantizer that is used in the spatial domain as well as in the frequency domain.
2 more elements
Dependencies are inferred from the claim text.