List your company

Video retrieval on UCF101

90.7Top-1 AccSigLIP2-g-opt
7.529.150.772.3Oct 27, 2020Sep 5, 2021Jul 16, 2022May 26, 2023Apr 3, 2024Feb 11, 2025Dec 22, 2025

Evaluation results

MethodLinks
SigLIP2-g-optV-Enc Params=1.1B, Resolution=384, Frame/ fps=8, Zero-shot Protocol=true2025.1290.7————
PE-GV-Enc Params=1.9B, Video Data=22M, Resolution=448, Frame/ fps=8, Zero-shot Protocol=true2025.1290.7————
PEAV-LV-Enc Params=0.5B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.1290.4————
PEAV-BV-Enc Params=0.4B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.1289.7————
PEAV-BV-Enc Params=0.4B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.1289.5————
PEAV-LV-Enc Params=0.5B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.1289.2————
PEAV-SV-Enc Params=0.3B, Video Data=124M, Resolution=336, Frame/ fps=30*, Zero-shot Protocol=true2025.1289.2————
InternVideo2V-Enc Params=1.0B, Video Data=102M, Resolution=224, Frame/ fps=8, Zero-shot Protocol=true2025.1288.8————
PEAV-SV-Enc Params=0.3B, Video Data=124M, Resolution=336, Frame/ fps=16, Zero-shot Protocol=true2025.1287.8————
PE-LV-Enc Params=0.3B, Video Data=22M, Resolution=336, Frame/ fps=8, Zero-shot Protocol=true2025.1287.1————
SigLIP2-L/16V-Enc Params=0.3B, Resolution=384, Frame/ fps=8, Zero-shot Protocol=true2025.1286.7————
LanguageBindV-Enc Params=0.3B, Video Data=10M, Resolution=224, Frame/ fps=16, Zero-shot Protocol=true2025.1281.3————
ImageBindV-Enc Params=0.6B, Video Data=3M, Resolution=224, Frame/ fps=16, Zero-shot Protocol=true2025.1277.1————
FAMEBackbone=R(2+1)D, Pre-train Dataset=Kinetics-4002021.0964.677.782.987.694.2
ASCNetBackbone=R3D-18, Pre-train Dataset=Kinetics-4002021.0958.976.382.287.593.4
GDTBackbone=R(2+1)D, Pre-train Dataset=Kinetics-4002021.0957.473.480.888.192.9
TCLRBackbone=R(2+1)D-18, Pre-training=UCF1012021.0156.972.27984.6—
TCLRBackbone=R3D-18, Pre-training=UCF1012021.0156.272.27985.3—
TCLRBackbone=C3D, Pre-training=UCF1012021.0148.667.675.582.5—
SSTLBackbone=R3D-18, Pre-training=Kinetics-4002021.0144.557.463.570—
CACLNet=C3D, Pre-trained=Kinetics4002022.0544.263.171.980.489.4
CACLNet=C3D2022.0543.261.169.978.288.2
MLRepBackbone=R3D-18, Pre-train Dataset=Kinetics-4002021.0941.56071.280.1—
CACLNet=R(2+1)D2022.0541.559.768.477.687.8
RSPNetArchitecture=ResNet-182020.1041.159.468.477.888.7
RSPNetBackbone=R3D-18, Venue=AAAI-21, Pre-training=UCF1012021.0141.159.468.477.8—
CACLNet=R3D2022.0541.159.267.375.285.9
MFOBackbone=R3D-18, Venue=ICCV-21, Pre-training=UCF1012021.0139.657.669.278—
STSBackbone=C3D, Venue=TPAMI-21, Pre-training=Kinetics-4002021.0139.159.268.877.6—
STSBackbone=R3D-18, Venue=TPAMI-21, Pre-training=Kinetics-4002021.0138.359.968.977.2—
STSBackbone=R(2+1)D-18, Venue=TPAMI-21, Pre-training=Kinetics-4002021.0138.158.968.177—
RSPNetArchitecture=C3D2020.103656.766.576.387.7
RSPNetBackbone=C3D, Venue=AAAI-21, Pre-training=UCF1012021.013656.766.576.3—
RSPNetNet=C3D2022.053656.766.576.387.7
PaceArchitecture=C3D2020.1031.949.759.268.980.2
Pace PredBackbone=C3D, Venue=ECCV-20, Pre-training=UCF1012021.0131.949.759.268.9—
Pace PredNet=C3D2022.0531.949.759.268.980.2
TempTransBackbone=R3D-18, Pre-train Dataset=Kinetics-4002021.0926.148.559.169.682.8
Temp TransBackbone=R3D-18, Venue=ECCV-20, Pre-training=UCF1012021.0126.148.559.169.6—
Buchler et al.Architecture=CaffeNet2020.1025.736.242.249.259.5
Pace PredBackbone=R(2+1)D-18, Venue=ECCV-20, Pre-training=UCF1012021.0125.642.751.361.3—
Pace PredNet=R(2+1)D2022.0525.642.751.361.374
Var. PSPBackbone=R3D-18, Venue=Access-21, Pre-training=UCF1012021.0124.641.951.362.7—
Pace PredBackbone=R3D-18, Venue=ECCV-20, Pre-training=UCF1012021.0123.838.146.456.6—
Pace PredNet=R3D2022.0523.838.146.456.669.8
CSJBackbone=R3D-18, Pre-training=Kinetics-4002021.0121.533.553.264.9—
MemDPCBackbone=R3D-18, Venue=ECCV-20, Pre-training=UCF1012021.0120.240.452.464.7—
OPNArchitecture=OPN2020.1019.928.73440.651.6
VCPArchitecture=R(2+1)D2020.1019.933.74250.564.4
VCPBackbone=R(2+1)D-18, Venue=AAAI-20, Pre-training=UCF1012021.0119.933.74250.5—
VCPBackbone=R3D-18, Venue=AAAI-20, Pre-training=UCF1012021.0118.633.642.553.5—
VCPBackbone=C3D, Venue=AAAI-20, Pre-training=UCF1012021.0117.331.54252.6—
DSMBackbone=C3D, Venue=AAAI-21, Pre-training=UCF1012021.0116.833.443.454.6—
DSMNet=C3D2022.0516.833.443.454.670.7
ClipOrderArchitecture=R3D2020.1014.130.34051.166.5
VCOPBackbone=R3D-18, Venue=CVPR-20, Pre-training=UCF1012021.0114.130.340.451.1—
VCOPNet=R3D2022.0514.130.34051.166.5
SpeedNetArchitecture=S3D-G2020.101328.137.549.565
SpeedNetBackbone=S3D-G, Pre-train Dataset=Kinetics-4002021.091328.137.549.565
VCOPBackbone=C3D, Venue=CVPR-20, Pre-training=UCF1012021.0112.5293950.6—
VCOPNet=C3D2022.0512.5293950.666.9
VCOPBackbone=R(2+1)D-18, Venue=CVPR-20, Pre-training=UCF1012021.0110.725.935.447.3—
VCOPNet=R(2+1)D2022.0510.725.935.447.363.9