List your company
ResearchFormerly WizWand

Video Quality Assessment on LIVE-VQC (SRCC, PLCC)

0.884SRCCSoft Ranking (Stage 1)
0.7280.76850.8090.8495Oct 11, 2022May 18, 2023Dec 24, 2023Jul 30, 2024Mar 7, 2025Oct 12, 2025May 20, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Soft Ranking (Stage 1)Training Stage=Stage 1, Evaluation Protocol=10-fold cross-validation2025.050.8840.894
KSVQE2024.020.8610.883
qstRectifiers=Spatial and Temporal2024.020.860.88
DOVERprotocol=supervised2026.040.860.875
DOVEREvaluation Protocol=10-fold cross-validation2025.050.860.875
KVQPre-training Dataset=LSVQ [56]2025.030.8590.879
KVQprotocol=supervised2026.040.8590.879
StarVQA+Pre-training Dataset=ImageNet [7], LSVQ [56], ...2025.030.8570.874
FastVQA2024.020.8530.873
DOVER2024.020.8530.872
FastVQA2024.020.8490.862
Fast-VQAPre-training Dataset=LSVQ [56]2025.030.8490.862
FastVQAprotocol=supervised2026.040.8490.865
FAST-VQAEvaluation Protocol=10-fold cross-validation2025.050.8490.865
SimpleVQAprotocol=supervised2026.040.8450.859
SimpleVQAEvaluation Protocol=10-fold cross-validation2025.050.8450.859
Doveraesthetic branch=false2024.020.8440.875
FasterVQAPre-training Dataset=LSVQ [56]2025.030.8430.858
MinimalisticVQAEvaluation Protocol=10-fold cross-validation2025.050.8420.854
Li222024.020.8410.839
DPC-VQAprotocol=few-shot2026.040.8410.853
Soft Ranking (Base)Training Stage=Base, Evaluation Protocol=10-fold cross-validation2025.050.840.853
Li et al.Pre-training Dataset=BID [6], LIVE [9], KonIQ-10k [15], SPAQ [8]2025.030.8340.842
BVQAprotocol=supervised2026.040.8340.842
qtRectifiers=Temporal2024.020.8330.851
DOVERInference Time (Sec)=0.047, Testing Protocol=Cross-dataset Testing2024.020.8320.855
PVQ2024.020.8270.837
PVQPre-training Dataset=PaQ-2-PiQ [55]2025.030.8270.837
QPT V2Evaluation Protocol=Cross-dataset2024.070.8270.853
PatchVQEvaluation Protocol=10-fold cross-validation2025.050.8270.837
FAST-VQATraining data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.8260.845
FAST-VQAMethod Category=Supervised VQA/IQA models2026.050.8260.845
VQTPre-training Dataset=ImageNet [7], Kinetics-400 [18]2025.030.8240.836
FAST-VQA2022.100.8230.844
FastVQAInference Time (Sec)=0.045, Testing Protocol=Cross-dataset Testing2024.020.8230.844
FastVQAEvaluation Protocol=Cross-dataset2024.070.8230.844
DisCoVQAPre-training Dataset=None2025.030.820.826
VQ-Jarvis2026.030.820.862
DOVERTraining data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.8170.84
DOVERMethod Category=Supervised VQA/IQA models2026.050.8170.84
BVQA-TCSVT-20222022.100.8160.824
BVQAEvaluation Protocol=Cross-dataset2024.070.8160.824
CONTRIQUEprotocol=few-shot2026.040.8150.822
FasterVQA2022.100.8130.837
CONVIQTprotocol=few-shot2026.040.8080.817
VQAThinkerMethod Category=Reasoning or reinforcement-learning based LMM evaluators2026.050.8080.847
qstInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.8060.844
FasterVQA-MTAMI=true2022.100.8030.826
qtInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.8030.837
LMM-PVQA (Soft Ranking Stage 3)Training data=PLVD-P1/ P2/ P3 (700k), Human Label=No, Setting=Zero-shot2025.050.8010.836
VersusQMethod Category=Reasoning or reinforcement-learning based LMM evaluators2026.050.8010.848
TLVQMPre-training Dataset=NA (pure handcraft)2025.030.7990.803
CSPTprotocol=few-shot2026.040.7990.819
TLVQM2024.020.7980.802
LMM-PVQA (Soft Ranking Stage 2)Training data=PLVD-P1/ P2 (600k), Human Label=No, Setting=Zero-shot2025.050.7980.833
LMM-PVQA (Soft Ranking Stage 1)Training data=PLVD-P1 (500k), Human Label=No, Setting=Zero-shot2025.050.7970.832
Full-res Swin-T feat., 32 x 4 frames2022.100.7940.809
Li22Inference Time (Sec)=27.632, Testing Protocol=Cross-dataset Testing2024.020.7930.811
FasterVQA-MSAMI=true2022.100.7910.818
LMM-PVQA (Hard Ranking)Training data=PLVD-P1 (500k), Human Label=No, Setting=Zero-shot2025.050.7910.82
VQ-Insight2026.030.790.835
VQ-InsightMethod Category=Reasoning or reinforcement-learning based LMM evaluators2026.050.790.835
FAST-VQA-M2022.100.7880.81
GSTVQA2024.020.7880.796
VQA2-ScorerMethod Category=Supervised VQA/IQA models2026.050.7850.83
BUONA-VISTATraining data=None, Human Label=No, Setting=Zero-shot2025.050.7840.794
Modular-VQA2026.030.7830.825
Q-AlignTraining data=fused [11, 17, 28, 40, 70], Human Label=Yes, Setting=Zero-shot2025.050.7830.819
Q-AlignMethod Category=Supervised VQA/IQA models2026.050.7830.819
PVQEvaluation Protocol=Cross-dataset, Patch usage=w/o patch2024.070.7810.781
Q-Align2026.030.7770.813
VQA^22026.030.7760.823
MinimalisticVQA(IX)Training data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.7750.821
MinimalisticVQAMethod Category=Supervised VQA/IQA models2026.050.7750.821
VSFA2024.020.7730.795
qsInference Time (Sec)=0.159, Testing Protocol=Cross-dataset Testing2024.020.7730.823
VSFAPre-training Dataset=None2025.030.7730.795
Q-Alignprotocol=supervised2026.040.7730.829
DOVER2026.030.7710.819
PVQ w/patch2022.100.770.807
PVQEvaluation Protocol=Cross-dataset, Patch usage=w/ patch2024.070.770.807
Fast-VQA2026.030.7690.815
Minimalist-VQA2026.030.7650.812
VideoLLaMA3 (Qwen2.5-7B)S2I-Tuned=Yes2025.060.7630.742
UCDAprotocol=few-shot2026.040.7620.77
SimpleVQAMethod Category=Supervised VQA/IQA models2026.050.7620.799
2BiVQAPre-training Dataset=ImageNet [7], KonIQ-10k [15]2025.030.7610.832
MinimalisticVQA(VII)Training data=LSVQ [70], Human Label=Yes, Setting=Zero-shot2025.050.7570.813
RAPIQUE2024.020.7540.786
VSFAInference Time (Sec)=11.109, Testing Protocol=Cross-dataset Testing2024.020.7530.795
StarVQAPre-training Dataset=LSVQ [56]2025.030.7530.809
VIDEVAL2024.020.7520.751
VIDEVALPre-training Dataset=NA (pure handcraft)2025.030.7520.751
SimpleVQAInference Time (Sec)=0.714, Testing Protocol=Cross-dataset Testing2024.020.7490.789
PVQ wo/patch2022.100.7470.776
SimpleVQA2024.020.740.775
RAPIQUEprotocol=supervised2026.040.740.764
LLaVA-OneVision (Vicuna-v1.1-7B)S2I-Tuned=Yes2025.060.7380.752
qbRectifiers=None2024.020.7370.786
VSFA2022.100.7340.772