Review: Lin DCC’16 — CNN for H.264 & HEVC Compressed Image Deblocking (Codec Filtering)

Recently, I have read some papers for post processing after decoding certain compressed image or video. These post processing steps help to increase the decoded image/video quality. This time, I would like to shortly review a very simple 2016 DCC paper for H.264 and HEVC convolutional neural network (CNN) based post processing. (Sik-Ho Tsang @ Medium)

Outline

1. Network Architecture
2. Network Training

1. Network Architecture

3-Layer Network Architecture

• The network is very simple, it just consists of 3 convolutional layers with the use of ReLU except the last layer.

2. Network Training

Network Training

• The CNN learns the connections between the lossy frame and the original one in an implicit way under different quantization parameters (QP), by minimizing the loss using stochastic gradient descent with the standard backpropagation.
• It is noted that with the lower (higher) QP, the higher (lower) image/video quality, the higher (lower) bitrate to compress the image/video.
• Finally, the CNN is implemented as a post-loop deblocking filter.
• Various experimental results demonstrate that the proposed method is able to significantly improve the quality of enhanced frames in terms of both objective and subjective criterions. (But there is no results shown in the paper.)

Reference

[2016 DCC] [Lin DCC’16]
Deep Convolutional Neural Network for Decompressed Video Enhancement

My Previous Reviews

Image Classification [LeNet] [AlexNet] [Maxout] [NIN] [ZFNet] [VGGNet] [Highway] [SPPNet] [PReLU-Net] [STN] [DeepImage] [SqueezeNet] [GoogLeNet / Inception-v1] [BN-Inception / Inception-v2] [Inception-v3] [Inception-v4] [Xception] [MobileNetV1] [ResNet] [Pre-Activation ResNet] [RiR] [RoR] [Stochastic Depth] [WRN] [Shake-Shake] [FractalNet] [Trimps-Soushen] [PolyNet] [ResNeXt] [DenseNet] [PyramidNet] [DRN] [DPN] [Residual Attention Network] [DMRNet / DFN-MR] [IGCNet / IGCV1] [MSDNet] [ShuffleNet V1] [SENet] [NASNet] [MobileNetV2]

Object Detection [OverFeat] [R-CNN] [Fast R-CNN] [Faster R-CNN] [MR-CNN & S-CNN] [DeepID-Net] [CRAFT] [R-FCN] [ION] [MultiPathNet] [NoC] [Hikvision] [GBD-Net / GBD-v1 & GBD-v2] [G-RMI] [TDM] [SSD] [DSSD] [YOLOv1] [YOLOv2 / YOLO9000] [YOLOv3] [FPN] [RetinaNet] [DCN]

Semantic Segmentation [FCN] [DeconvNet] [DeepLabv1 & DeepLabv2] [CRF-RNN] [SegNet] [ParseNet] [DilatedNet] [DRN] [RefineNet] [GCN] [PSPNet] [DeepLabv3] [LC] [FC-DenseNet] [IDW-CNN] [SDN]

Biomedical Image Segmentation [CUMedVision1] [CUMedVision2 / DCAN] [U-Net] [CFS-FCN] [U-Net+ResNet] [MultiChannel] [V-Net] [3D U-Net] [M²FCN] [SA] [QSA+QNT] [3D U-Net+ResNet]

Instance Segmentation [SDS] [Hypercolumn] [DeepMask] [SharpMask] [MultiPathNet] [MNC] [InstanceFCN] [FCIS]

Super Resolution [SRCNN] [FSRCNN] [VDSR] [ESPCN] [RED-Net] [DRCN] [DRRN] [LapSRN & MS-LapSRN] [SRDenseNet]

Human Pose Estimation [DeepPose] [Tompson NIPS’14] [Tompson CVPR’15] [CPM]

Codec Post-Processing [ARCNN] [Lin DCC’16] [Li ICME’17]