List your company
ResearchFormerly WizWand

Video Quality Assessment on KoNViD-1k

0.922SROCCKSVQE
0.785760.821130.85650.89187Jan 26, 2021Dec 9, 2021Oct 22, 2022Sep 5, 2023Jul 18, 2024May 31, 2025Apr 14, 2026
Updated 5mo ago

Evaluation results

MethodLinks
KSVQE2024.020.922——0.921——
Soft Ranking (Stage 1)Training Stage=Stage 1, Evaluation Protocol=10-fold cross-validation2025.050.911——0.908——
KVQPre-training Dataset=LSVQ [56]2025.030.909——0.915——
DOVERprotocol=supervised2026.040.909——0.906——
KVQprotocol=supervised2026.040.909——0.915——
Doveraesthetic branch=false2024.020.908——0.91——
DOVEREvaluation Protocol=10-fold cross-validation2025.050.908——0.91——
qstRectifiers=Spatial and Temporal2024.020.901——0.905——
Soft Ranking (Base)Training Stage=Base, Evaluation Protocol=10-fold cross-validation2025.050.896——0.877——
Q-ALIGN (1fps) + FAST-VQATraining Set=LSVQtrain, IQA Pre-training=false, fps=12023.120.895——0.897——
FasterVQAPre-training Dataset=LSVQ [56]2025.030.895——0.898——
LMM-PVQA (Soft Ranking Stage 3)Training data=PLVD-P1/ P2/ P3 (700k), Human Label=No, Setting=Zero-shot2025.050.894——0.893——
FastVQA2024.020.893——0.887——
LMM-PVQA (Soft Ranking Stage 2)Training data=PLVD-P1/ P2 (600k), Human Label=No, Setting=Zero-shot2025.050.893——0.892——
DOVER2024.020.892——0.9——
qtRectifiers=Temporal2024.020.892——0.891——
FastVQA2024.020.891——0.892——
Fast-VQAPre-training Dataset=LSVQ [56]2025.030.891——0.892——
VQ-Jarvis2026.030.891——0.899——
FastVQAprotocol=supervised2026.040.891——0.892——
FAST-VQAEvaluation Protocol=10-fold cross-validation2025.050.891——0.892——
LMM-PVQA (Soft Ranking Stage 1)Training data=PLVD-P1 (500k), Human Label=No, Setting=Zero-shot2025.050.891——0.888——
DOVER2026.030.89——0.883——
MinimalisticVQAEvaluation Protocol=10-fold cross-validation2025.050.889——0.89——
LMM-PVQA (Hard Ranking)Training data=PLVD-P1 (500k), Human Label=No, Setting=Zero-shot2025.050.886——0.869——
DOVERInference Time (Sec)=0.047, Testing Protocol=Cross-dataset Testing2024.020.884——0.883——
DOVERTraining Set=LSVQtrain, IQA Pre-training=false, Components=aesthetic branch + FAST-VQA2023.120.883——0.884——
DPC-VQAprotocol=few-shot2026.040.883——0.891——
qstInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.878——0.884——
Modular-VQA2026.030.878——0.884——
StarVQA+Pre-training Dataset=ImageNet [7], LSVQ [56], ...2025.030.876——0.881——
Q-AlignTraining data=fused [11, 17, 28, 40, 70], Human Label=Yes, Setting=Zero-shot2025.050.876——0.878——
VQ-Insight2026.030.875——0.884——
DOVERTraining data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.874——0.869——
qsRectifiers=Spatial2024.020.868——0.878——
QPT V2Evaluation Protocol=Cross-dataset2024.070.866——0.865——
Q-ALIGN (1fps)Training Set=LSVQtrain, IQA Pre-training=false, fps=12023.120.865——0.877——
Q-AlignModel Type=Specialized VQA Model2026.020.865——0.877——
Q-Align2026.030.865——0.876——
Q-AlignEvaluation Protocol=Cross-dataset2024.070.865——0.877——
Q-Alignprotocol=supervised2026.040.865——0.877——
qbRectifiers=None2024.020.864——0.875——
FasterVQA2022.100.863——0.863——
MinimalisticVQA(IX)Training data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.862——0.859——
SimpleVQAEvaluation Protocol=Cross-dataset2024.070.86—————
FAST-VQA2022.100.859——0.855——
FAST-VQATraining Set=LSVQtrain, IQA Pre-training=false2023.120.859——0.855——
FastVQAInference Time (Sec)=0.045, Testing Protocol=Cross-dataset Testing2024.020.859——0.855——
Fast-VQA2026.030.859——0.857——
Minimalist-VQA2026.030.859——0.861——
FastVQAEvaluation Protocol=Cross-dataset2024.070.859——0.855——
FAST-VQATraining data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.859——0.854——
VQTPre-training Dataset=ImageNet [7], Kinetics-400 [18]2025.030.858——0.868——
SimpleVQAType=VQA2022.040.856——0.86——
SimpleVQA2024.020.856——0.86——
SimpleVQAprotocol=supervised2026.040.856——0.86——
SimpleVQAEvaluation Protocol=10-fold cross-validation2025.050.856——0.86——
FasterVQA-MSAMI=true2022.100.852——0.854——
DisCoVQATraining Set=LSVQtrain, IQA Pre-training=false2023.120.851——0.853——
qtInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.851——0.853——
CONVIQTprotocol=few-shot2026.040.851——0.849——
DisCoVQAPre-training Dataset=None2025.030.847——0.847——
FasterVQA-MTAMI=true2022.100.846——0.849——
qsInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.845——0.856——
CONTRIQUEprotocol=few-shot2026.040.844——0.842——
Li22Inference Time (Sec)=27.632, Testing Protocol=Cross-dataset Testing2024.020.843——0.835——
MinimalisticVQA(VII)Training data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.843——0.841——
BVQA-2022Attack Status=Before Attack2023.090.8427——0.8469——
StarVQAPre-training Dataset=LSVQ [56]2025.030.842——0.849——
FAST-VQA-M2022.100.841——0.832——
SimpleVQATraining Set=LSVQtrain, IQA Pre-training=true2023.120.84——0.834——
BVQATraining Set=LSVQtrain, IQA Pre-training=true2023.120.839——0.83——
Li222024.020.839——0.83——
BVQAEvaluation Protocol=Cross-dataset2024.070.839——0.83——
Li et al.Type=VQA2022.040.836——0.834——
BVQA-TCSVT-20222022.100.834——0.837——
Li et al.Pre-training Dataset=BID [6], LIVE [9], KonIQ-10k [15], SPAQ [8]2025.030.834——0.836——
BVQAprotocol=supervised2026.040.834——0.836——
VQA^22026.030.833——0.844——
SimpleVQAInference Time (Sec)=0.714, Testing Protocol=Cross-dataset Testing2024.020.826——0.82——
SSL-VQAprotocol=few-shot2026.040.826——0.828——
Full-res Swin-T feat., 32 x 4 frames2022.100.825——0.828——
qbInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.822——0.836——
2BiVQAPre-training Dataset=ImageNet [7], KonIQ-10k [15]2025.030.815——0.835——
GSTVQA2024.020.814——0.825——
RAPIQUEprotocol=supervised2026.040.811——0.819——
VSFAInference Time (Sec)=11.109, Testing Protocol=Cross-dataset Testing2024.020.81——0.811——
TiVQAAttack Status=Before Attack2023.090.8046——0.8377——
RAPIQUEbackbone=ResNet-50, sampling_rate=2 fps2021.010.8031——0.81750.3623—
RAPIQUEType=VQA2022.040.803——0.818——
RAPIQUE2024.020.803——0.817——
ResNet50Type=IQA2022.040.802——0.81——
ASOTraining Strategy=Analytic Score Optimization2026.020.801——0.835——
CSPTprotocol=few-shot2026.040.801——0.806——
MDTVSFAAttack Status=Before Attack2023.090.8003——0.8074——
VSFA2024.020.794——0.799——
VSFAprotocol=supervised2026.040.794——0.799——
VSFAEvaluation Protocol=10-fold cross-validation2025.050.794——0.799——
SimpleVQA2024.020.792——0.798——
PVQ w/patch2022.100.791——0.795——