HyperAI超神经

首页资讯最新论文教程数据集百科 SOTA LLM 模型天梯 GPU 天梯顶会

中文

HyperAI超神经

Jpeg Artifact Correction On Live1 Quality 20 1

评估指标

PSNR

PSNR-B

SSIM

评测结果

各个模型在此基准测试上的表现结果

模型名称	PSNR	PSNR-B	SSIM	Paper Title	Repository
MWCNN	32.04	31.83	0.8989	Multi-level Wavelet-CNN for Image Restoration
RED30	31.73	-	-	Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections
FBCNN	32.13	31.57	0.889	Towards Flexible Blind JPEG Artifacts Removal
S-Net	31.83	31.76	0.8975	S-Net: A Scalable Convolutional Neural Network for JPEG Compression Artifact Reduction	-
DnCNN-3	31.59	-	-	Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
Residual Dense Network +	32.1	-	0.8886	Residual Dense Network for Image Restoration
ARCNN	31.29	31.37	0.8891	Compression Artifacts Reduction by a Deep Convolutional Network
MemNet	31.83	31.74	0.8970	MemNet: A Persistent Memory Network for Image Restoration
DPW-SDNet	31.69	31.60	0.8891	DPW-SDNet: Dual Pixel-Wavelet Domain Deep CNNs for Soft Decoding of JPEG-Compressed Images	-
Hi-IR	32.31	-	0.8938	Hierarchical Information Flow for Generalized Efficient Image Restoration	-
IDCN	32.09	32.00	0.9006	Implicit Dual-domain Convolutional Network for Robust Color Image Compression Artifact Reduction	-
QGAC	31.86	31.27	0.901	Quantization Guided JPEG Artifact Correction

0 of 12 row(s) selected.