Video encoding apparatus and video encoding method that perform filtering operation during video encoding process

Patent metadata

Field	Value
Publication number	US-10477235-B2
Application number	US-201715646096-A
Country	US
Kind code	B2
Filing date	Jul 11, 2017
Priority date	Jul 11, 2017
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

A video encoding apparatus and a video encoding method are provided. The video encoding apparatus includes an integer-pel motion estimation (IME) circuit, a temporal noise reduction (TNR) circuit, a fractional-pel motion estimation (FME) circuit and an encoding circuit. The IME circuit provides the first motion vector, the error value and the co-located error value of the current block in the current frame to the TNR circuit. By using the first motion vector, the error value and the co-located error value of the current block in the current frame, the TNR circuit performs the temporal filtering process on the current block in an original image data to produce a denoised image data to the FME circuit.

First claim

What is claimed is:

1. A video encoding apparatus, comprising:
an integer-pel motion estimation circuit, configured to perform an integer-pel motion estimation operation on an original image data, wherein the integer-pel motion estimation circuit generates a first motion vector, an error value and a co-located error value for a current block during the integer-pel motion estimation operation, the error value corresponds to the first motion vector, and the co-located error value is a difference between the current block in a current frame and a co-located block in a reference frame;
a temporal noise reduction circuit, coupled to the integer-pel motion estimation circuit to receive the first motion vector, the error value and the co-located error value of the current block, and configured to perform a temporal filtering process on the current block in the original image data by using the first motion vector, the error value and the co-located error value of the current block in the current frame to obtain a denoised image data;
a fractional-pel motion estimation circuit, coupled to the temporal noise reduction circuit to receive the denoised image data, the first motion vector, the error value and the co-located error value, configured to perform a fractional-pel motion estimation operation on the denoised image data to at least produce a second motion vector of the current block according to the first motion vector, the error value and the co-located error value; and
an encoding circuit, coupled to the fractional-pel motion estimation circuit to receive the second motion vector, and configured to perform a video encoding operation.

2. The video encoding apparatus according to claim 1 , wherein the temporal noise reduction circuit performs moving object detection for detecting whether the current block in the current frame is a moving object to obtain a detection result by using the first motion vector, the error value and the co-located error value of the current block in the current frame and using a motion vector of the co-located block in the reference frame, and the temporal noise reduction circuit dynamically sets a temporal filtering parameter of the temporal filtering process according to the detection result.

3. The video encoding apparatus according to claim 2 , wherein the temporal noise reduction circuit analyzes a current motion characteristic of the current block in the current frame and a reference motion characteristic of the co-located block in the reference frame by using the first motion vector, the error value and the co-located error value of the current block in the current frame and using the motion vector of the co-located block in the reference frame, and the temporal noise reduction circuit determines whether the current block in the current frame is the moving object according to the current motion characteristic and the reference motion characteristic.

4. The video encoding apparatus according to claim 2 , wherein the temporal filtering parameter of the temporal filtering process is decreased when the current block is determined as the moving object.

5. The video encoding apparatus according to claim 1 , wherein the temporal filtering process includes a first temporal filtering process and a second temporal filtering process, the temporal noise reduction circuit performs the first temporal filtering process on the current block in the original image data to obtain a first denoised image data, the temporal noise reduction circuit performs the second temporal filtering process on the current block in the original image data to obtain a second denoised image data, the temporal noise reduction circuit blends the first denoised image data and the second denoised image data to obtain the denoised image data according to a blending weight, and the temporal noise reduction circuit decreases the blending weight when the current block is determined as a moving object.

6. The video encoding apparatus according to claim 1 , wherein the temporal noise reduction circuit comprises:
a parameter circuit, coupled to the integer-pel motion estimation circuit to receive the first motion vector of the current block, detecting whether the current block in the current frame is a moving object by using a motion vector of the co-located block in the reference frame and using the first motion vector of the current block in the current frame to obtain a detection result, and dynamically setting a temporal filtering parameter according to the detection result; and
a filter circuit, coupled to the parameter circuit to receive the temporal filtering parameter, and performing the temporal filtering process on the current block in the original image data by using the temporal filtering parameter to obtain the denoised image data.

7. The video encoding apparatus according to claim 6 , wherein the parameter circuit comprises:
an adjustment circuit, coupled to the integer-pel motion estimation circuit to receive the first motion vector, the error value and the co-located error value of the current block, and determining whether to set the first motion vector of the current block to a zero vector according to the error value and the co-located error value of the current block; and
a parameter generation circuit, coupled to the adjustment circuit to receive the first motion vector, detecting whether the current block in the current frame is the moving object by using the motion vector of the co-located block in the reference frame and the first motion vector of the current block in the current frame to obtain the detection result, and dynamically setting the temporal filtering parameter according to the detection result.

8. The video encoding apparatus according to claim 6 , wherein the current block comprises a plurality of sub-blocks, and the parameter circuit comprises:
an adjustment circuit, coupled to the integer-pel motion estimation circuit to receive the first motion vector, the error value and the co-located error value of the current block, wherein the adjustment circuit accumulates motion vectors of a plurality of sub-blocks of the co-located block in the reference frame to obtain reference motion characteristic of the co-located block in the reference frame, the adjustment circuit accumulates motion vectors of the sub-blocks of the current block in the current frame to obtain current motion characteristic of the current block in the current frame, and determining whether to set the current motion characteristic of the current block to zero according to the error value and the co-located error value of the current block; and
a parameter generation circuit, coupled to the adjustment circuit to receive the reference motion characteristic and the current motion characteristic, wherein the parameter generation circuit detects whether the current block in the current frame is the moving object by using the reference motion characteristic and the current motion characteristic to obtain the detection result, and dynamically setting the temporal filtering parameter according to the detection result.

9. The video encoding apparatus according to claim 1 , wherein the temporal noise reduction circuit comprises:
a parameter circuit, coupled to the integer-pel motion estimation circuit to receive the first motion vector of the current block, detecting whether the current block in the current frame is a moving object by using a motion vector of the co-located block in the reference frame and the first motion vector of the current block in the current frame to obtain a detection result, and dynamically setting a first temporal filtering parameter and a second temporal filtering parameter according to the detection result; and
a filter circuit, coupled to the

Assignees

• Novatek Microelectronics Corp

Inventors

• Peng Ting
• Jou Fan-Di
• Huang Xin

Classifications

• H04N19/521

  for estimating the reliability of the determined motion vectors or motion vector field, e.g. for smoothing the motion vector field or for correcting motion vectors · CPC title

• H04N19/523 Primary

  with sub-pixel accuracy · CPC title

• H04N19/23

  with coding of regions that are present throughout a whole video segment, e.g. sprites, background or mosaic · CPC title

• H04N19/82

  involving filtering within a prediction loop · CPC title