HyperAI초신경

홈 뉴스 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Image Super Resolution On Set5 4X Upscaling

평가 지표

PSNR

SSIM

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	PSNR	SSIM	Paper Title	Repository
MaIR+	33.14	0.9058	MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration
SRCNN	30.49	0.8628	Image Super-Resolution Using Deep Convolutional Networks
CPAT+	33.24	0.9071	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
ML-CrAIST	32.36	0.8984	ML-CrAIST: Multi-scale Low-high Frequency Information-based Cross black Attention with Image Super-resolving Transformer
SRGAN	29.4	0.8472	Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
EDSR	32.46	0.8968	Enhanced Deep Residual Networks for Single Image Super-Resolution
MaIR	32.93	0.9045	MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration
ML-CrAIST-Li	32.15	0.8962	ML-CrAIST: Multi-scale Low-high Frequency Information-based Cross black Attention with Image Super-resolving Transformer
CPAT	33.19	0.9069	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
HAT-L	33.30	0.9083	Activating More Pixels in Image Super-Resolution Transformer
Hi-IR-L	33.22	0.9103	Hierarchical Information Flow for Generalized Efficient Image Restoration	-
HMA†	33.38	0.9089	HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution

0 of 12 row(s) selected.