Review: RHCNN — Residual Highway Convolutional Neural Network (Codec Filtering)

Outperforms ARCNN

Outline

1. What is In-loop Filter

The pipeline of hybrid video coding

2. Choices of Basic Unit

Types of highway units. (a) Proposed, (b) Constant Scaling, (c) Dropout Shortcut, (d) Convolution Shortcut (X: concatenation, +: addition)
Gain of PSNR (dB) for (a) Proposed, (b) Constant Scaling, (c) Dropout Shortcut, (d) Convolution Shortcut

3. Entire Network Architecture

(a) 13-layer plain network. (b) 13-layer residual network. (c) 13-layer RHCNN.
Gain of PSNR (dB) for (a) 13-layer plain network. (b) 13-layer residual network. (c) 13-layer RHCNN.
Different Network Depths (Different Colors) Along Iterations.
Gain of PSNR (dB) for RHCNN with Various Depth

4. Experimental Results

Training and Validation Sets
Results for I frames
(a) Groundtruth, (b) Original HEVC, © HEVC with ARCNN, (d) HEVC with RHCNN
Running Time in Seconds Per Frame

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store