List your company
ResearchFormerly WizWand

Video Quality Assessment on CVD 2014 (test)

0.896SRCCFasterVQA
0.643280.708890.77450.84011Aug 19, 2021Oct 27, 2021Jan 5, 2022Mar 16, 2022May 24, 2022Aug 2, 2022Oct 11, 2022
Updated 8mo ago

Evaluation results

MethodLinks
FasterVQAResolution Range=480P - 720P, Efficiency=4X than FAST-VQA2022.100.8960.904
FAST-VQAGroup=Ours, Protocol=Finetune2022.070.8910.903
FAST-VQAResolution Range=480P - 720P2022.100.8910.903
FAST-VQA-MGroup=Ours, Protocol=Finetune2022.070.8770.892
FAST-VQA-MResolution Range=480P - 720P2022.100.8770.892
PVQTraining Setting=Mixed Databases2021.080.87370.8793
BVQA-TCSVT-2022Group=Existing Fixed Deep, Resolution Range=480P - 720P2022.100.8720.869
FAST-VQA w/o VQ-representationsGroup=Ours, Protocol=Finetune2022.070.8710.888
VSFAGroup=Existing Fixed Deep, Protocol=Finetune2022.070.870.868
VSFAGroup=Existing Fixed Deep, Resolution Range=480P - 720P2022.100.870.868
Full-res Swin-T featuresGroup=Baseline Features, Protocol=Finetune2022.070.8680.87
Full-res Swin-TResolution Range=480P - 720P, Features=Full-resolution Swin-T2022.100.8680.87
CNN+TLVQMGroup=Ensemble C+D, Protocol=Finetune2022.070.8630.88
CNN+TLVQMGroup=Ensemble C+D, Resolution Range=480P - 720P2022.100.8630.88
Li et al.Model Type=Supervised Pretraining2022.060.8620.882
Li et al.Model Type=Supervised Pretraining2022.060.8620.882
CONVIQTModel Type=Unsupervised Pretraining2022.060.8580.837
CONVIQTModel Type=Unsupervised Pretraining2022.060.8580.837
VSFAModel Type=Supervised Pretraining2022.060.850.869
VSFAModel Type=Supervised Pretraining2022.060.850.869
Proposed-LSTraining Setting=Mixed Databases, Scaling=Linear Re-scaling (LS)2021.080.84060.8519
Resnet-50Model Type=Supervised Pretraining, Pretrained=ImageNet2022.060.8350.838
Resnet-50Model Type=Supervised Pretraining2022.060.8350.838
MDTVSFATraining Setting=Mixed Databases2021.080.83260.8347
GSTVQAModel Type=Supervised Pretraining2022.060.8310.844
GSTVQAModel Type=Supervised Pretraining2022.060.8310.844
GST-VQAGroup=Existing Fixed Deep, Protocol=Finetune2022.070.8310.844
GST-VQAGroup=Existing Fixed Deep, Resolution Range=480P - 720P2022.100.8310.844
TLVQMGroup=Existing Classical, Protocol=Finetune2022.070.830.85
TLVQMGroup=Existing Classical, Resolution Range=480P - 720P2022.100.830.85
CONTRIQUEModel Type=Unsupervised Pretraining2022.060.8260.836
CONTRIQUEModel Type=Unsupervised Pretraining2022.060.8260.836
VIDEVALModel Type=Traditional/Handcrafted Features2022.060.8140.832
VIDEVALModel Type=Traditional/Handcrafted Features2022.060.8140.832
RAPIQUEModel Type=Traditional/Handcrafted Features2022.060.8070.823
RAPIQUEModel Type=Traditional/Handcrafted Features2022.060.8070.823
VBLIINDSModel Type=Traditional/Handcrafted Features2022.060.7950.806
VBLIINDSModel Type=Traditional/Handcrafted Features2022.060.7950.806
BRISQUEModel Type=Traditional/Handcrafted Features2022.060.790.804
BRISQUEModel Type=Traditional/Handcrafted Features2022.060.790.804
TLVQMModel Type=Traditional/Handcrafted Features2022.060.7790.79
TLVQMModel Type=Traditional/Handcrafted Features2022.060.7790.79
CSPTModel Type=Unsupervised Pretraining, Evaluation Protocol=linear evaluation2022.060.6530.691
CSPTModel Type=Unsupervised Pretraining, Evaluation Protocol=Linear evaluation2022.060.6530.691