Video Super Resolution On Vid4 4X Upscaling

Métriques

PSNR

SSIM

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	PSNR	SSIM	Paper Title	Repository
MIA-VSR	28.20	0.8507	Video Super-Resolution Transformer with Masked Inter&Intra-Frame Attention	-
VideoGigaGAN	27.04	-	VideoGigaGAN: Towards Detail-rich Video Super-Resolution	-
IconVSR	27.39	0.8279	BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond	-
EvTexture+	29.78	0.8983	EvTexture: Event-driven Texture Enhancement for Video Super-Resolution	-
BasicVSR	27.24	0.8251	BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond	-
BRCN	24.43	0.662	Bidirectional Recurrent Convolutional Networks for Multi-Frame Super-Resolution	-
RVRT	27.99	0.8462	Recurrent Video Restoration Transformer with Guided Deformable Attention	-
SATeCo	27.44	0.8420	Learning Spatial Adaptation and Temporal Coherence in Diffusion Models for Video Super-Resolution	-
ESPCN	25.06	0.7394	Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network	-
DBPN	25.37	0.737	Deep Back-Projection Networks For Super-Resolution	-
DRDVSR	25.88	0.774	Detail-revealing Deep Video Super-resolution	-
RBPN/6-PF	27.12	0.8180	Recurrent Back-Projection Network for Video Super-Resolution	-
CFD-PSRT	28.18	0.8503	Collaborative Feedback Discriminative Propagation for Video Super-Resolution	-
IART	28.26	0.8517	Enhancing Video Super-Resolution via Implicit Resampling-based Alignment	-
VESPCN	25.35	0.7557	Real-Time Video Super-Resolution with Spatio-Temporal Networks and Motion Compensation	-
BasicVSR++	27.79	0.8400	BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment	-
PSRT-recurrent	28.07	0.8485	Rethinking Alignment in Video Super-Resolution Transformers	-
VRT	27.93	0.8425	VRT: A Video Restoration Transformer	-
TecoGAN⊖	25.89	-	Learning Temporal Coherence via Self-Supervision for GAN-based Video Generation	-
SOF-VSR	26.01	0.771	Learning for Video Super-Resolution through HR Optical Flow Estimation	-

0 of 27 row(s) selected.