Sorry, you need to enable JavaScript to visit this website.

A Rate Control Scheme for HEVC Intra Coding Using Convolution Neural Network (CNN)

Citation Author(s):
Xin Lu, Bixing Zhou, Xuesong Jin and Graham Martin
Submitted by:
XIN LU
Last updated:
19 March 2020 - 1:56am
Document Type:
Poster
Document Year:
2020
Event:
Presenters:
Xin Lu
Paper Code:
DCC-128
Categories:
 

High-Efficiency Video Coding (HEVC) is the latest video coding standard which is developed by the Joint Collaborative Team on Video Coding (JCT-VC). To guarantee successful transmission and to make the best use of available network resources, an effective rate control mechanism plays a critical role in video coding standards. The coding performance can be maximised through the appropriate allocation of bits under the constraints of a total bit rate budget and the buffer size. The existing rate control strategies generally determine the rate control parameters according to the previous encoding results. However, there is no priori-knowledge available when the first frame of a video sequence is encoded or scene changes occur. Therefore, it is challenging to choose appropriate rate control parameters in such cases. In order to address this issue, we investigated the deep learning oriented methods and proposed a Convolutional Neural Network (CNN)-based rate control scheme for HEVC intra-coding.
As the CNN has the advantage of local perception and the CTU’s Quantisation Parameter (QP) value is determined at the CTU level rate control process, we applied the CNN at the CTU level as well to achieve an accurate QP prediction. An improved QP determination model (R-QP model) involving only two model coefficients (µ and ν) was suggested to simplify the QP calculation process. CNN was used to predict the model coefficients, namely µ and ν. Our CNN has four convolution layers, two pooling layers and three fully connected layers. Each of the convolution layers contains a Rectified Linear Unit (ReLU). The last fully connected layer outputs the predicted values of µ and ν. The images from the UCID dataset and the RAISE dataset were used to train the CNN. In order to obtain a universal model, none of the frame in the HEVC common test sequences was employed as a training image.
In order to achieve a reasonable allocation of bit budget, a CNN-based framework was used to predict the bit consumption for each CTU according to the entire picture content it contains. At the training stage, the network took the original images and the QP values as the inputs, and the corresponding actual bit consumptions were used as the labels. Once the CNN coefficients have been obtained through training, the bit consumption for each CTU under different QP values can be directly predicted from the original images. The predicted bit consumption was subsequently used to readjust the bit budget for each CTU to achieve a more reasonable bit allocation.
We implemented the proposed CNN-based rate control scheme in the HEVC reference software HM 16.9. As the priori-knowledge about the previous encoding results has not been considered in the proposed algorithm, only the first frame of each video sequence was encoded. The experimental results showed that our proposed scheme performs better than the default rate control implementation in the reference software of HEVC, namely the BDBR is reduced by 1.33% on average while maintaining the same reconstructed picture quality. Compared with the state-of-the-art CNN-based rate control algorithm proposed by Li, our schemes achieved better coding efficiency. It can be concluded that the proposed R-QP model is effective in describing the rate-distortion relationship in HEVC, and the CNN-based bit allocation scheme is beneficial for making the best of every bit.

up
0 users have voted: