Slice level reference picture reconstruction
AssigneeAgora Lab, Inc.
InventorsSheng Zhong, Wei Dai, Zesen Zhuang
Life of this patent
Estimating the term…
Abstract
A method for encoding a video stream, including encoding, by a processor, a first slice of a first picture of the video stream; receiving, from a decoder, a feedback message indicative of having received the first slice of the first picture; and in response to receiving the feedback message indicative of having received the first slice of the first picture, setting the first slice of the first picture as a reference slice for encoding a second picture of the video stream, wherein the second picture is encoded after receiving the feedback message.
Claims15
A method for encoding a video stream, comprising:
- encoding, by a processor, a first slice of a first picture of the video stream;
- receiving, from a decoder, a feedback message indicative of having received the first slice of the first picture;
- and
- in response to receiving the feedback message indicative of having received the first slice of the first picture:
- based on a determination that the first picture has not been set as a reference picture, setting the first picture as a new long-term reference (LTR) picture for encoding a second picture of the video stream,
- setting the first slice of the first picture as a reference slice for encoding the second picture, wherein the second picture is encoded after receiving the feedback message, and the first slice of the first picture is set as a valid reference slice, wherein a set of reference pictures used for predicting the second picture comprises each picture associated with at least one valid reference slice.
A method for decoding an encoded video stream, comprising:
- receiving, from an encoder, data associated with a first slice of a first picture of the encoded video stream;
- based on a determination that all data required for decoding the first slice of the first picture has been received, sending, to the encoder, a feedback message indicative of having received the first slice of the first picture, and decoding, by a processor, the first slice of the first picture;
- receiving, from the encoder, data associated with a first slice of a second picture of the encoded video stream with reference to the first slice of the first picture;
- based on a determination that the first picture has not been set as a reference picture and in response to receiving, from the encoder, data associated with the first slice of the second picture with reference to the first slice of the first picture, setting the first picture as a new reference picture for decoding the second picture of the encoded video stream;
- and
- setting the first slice of the first picture as a reference slice for decoding the second picture of the encoded video stream, wherein the first picture is set as a new long-term reference (LTR) picture, and the first slice of the first picture is set as a valid reference slice, wherein a set of reference pictures used for predicting the second picture comprises each picture associated with at least one valid reference slice and a fourth picture decoded immediately prior to decoding the second picture.
An apparatus for decoding an encoded video stream, comprising:
- a non-transitory memory;
- and
- a processor, wherein the non-transitory memory includes instructions executable by the processor to:
- receive, from an encoder, data associated with a first slice of a first picture of the encoded video stream;
- based on a determination that all data required for decoding the first slice of the first picture has been received, send, to the encoder, a feedback message indicative of having received the first slice of the first picture, and decode, by a processor, the first slice of the first picture;
- and
- receive, from the encoder, data associated with a first slice of a second picture of the encoded video stream with reference to the first slice of the first picture;
- based on a determination that the first picture has not been set as a reference picture and in response to receiving, from the encoder, data associated with the first slice of the second picture with reference to the first slice of the first picture, setting the first picture as a new reference picture for decoding the second picture of the encoded video stream;
- and
- set the first slice of the first picture as a reference slice for decoding the second picture of the encoded video stream, wherein the first picture is set as a new long-term reference (LTR) picture, and the first slice of the first picture is set as a valid reference slice, wherein a set of reference pictures used for predicting the second picture comprises each picture associated with at least one valid reference slice and a fourth picture decoded immediately prior to decoding the second picture.
Dependencies are inferred from the claim text.