List your company
ResearchFormerly WizWand

Transition Video Question Answering on TGIF-QA (test)

97.6AccuracyVGT
77.00882.35487.793.046Mar 14, 2022May 11, 2022Jul 8, 2022Sep 5, 2022Nov 2, 2022Dec 30, 2022Feb 27, 2023
Updated 8mo ago

Evaluation results

MethodLinks
VGTPretrain Dataset=None, Pretrain Size=None, Text Encoder=BERT2023.0297.6
CoVGTPretrain Dataset=None, Pretrain Size=None, Text Encoder=RoBERTa2023.0297.6
VGT (PT)Pretrain Dataset=WV, Pretrain Size=0.18M, Text Encoder=BERT2023.0297.2
MERLOTPretrain Dataset=YT, CC, Pretrain Size=183M, Text Encoder=BERT2023.0296.2
CoVGT (PT)Pretrain Dataset=WV, Pretrain Size=0.18M, Text Encoder=RoBERTa2023.0296.2
All-in-one-B [384]Nets=CE, Params=110M, PT Samples=3.72M, Frames=3, Resolution=384x3842022.0395.1
All-in-one-B *Nets=CE, Params=110M, PT Samples=9.72M, Frames=3, Resolution=224x224, Note=Pre-training with additional YT-Temporal 180M2022.0394.7
All-in-one-BNets=CE, Params=110M, PT Samples=3.72M, Frames=3, Resolution=224x2242022.0394.3
All-in-one-BNets=CE, Params=110M, PT Samples=3.72M, Frames=1, Resolution=224x2242022.0394.2
VIOLETNets=T+V+CE, Params=198M, PT Samples=5.5M, Frames=16, Resolution=224x2242022.0393.6
All-in-one-SNets=CE, Params=33M, PT Samples=3.72M, Frames=3, Resolution=224x2242022.0392.7
MHNPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0290.8
ClipBERTPretrain Dataset=VG, COCO, Pretrain Size=None, Text Encoder=BERT2023.0287.8
ClipBERTNets=T+V+CE, Params=137M, PT Samples=5.6M, Frames=1x1, Resolution=224x2242022.0387.5
MASNPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0287.4
PGATPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0285.7
HQGAPretrain Dataset=None, Pretrain Size=None, Text Encoder=BERT2023.0285.6
SiaSReaPretrain Dataset=VG, COCO, Pretrain Size=None, Text Encoder=BERT2023.0285.3
All-in-one-TiNets=CE, Params=12M, PT Samples=3.72M, Frames=3, Resolution=224x2242022.0383.5
HOSTRPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0283
B2APretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0282.6
HAIRPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0282.3
HCRNNets=T+V+LSTM, Frames=16, Resolution=224x2242022.0381.4
HCRNPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0281.4
LGCNPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0281.1
QueSTNets=T+V+LSTM, Frames=16, Resolution=224x2242022.0381
HGAPretrain Dataset=None, Pretrain Size=None, Text Encoder=GloVe2023.0281
HeterogeneousNets=T+V+LSTM, Frames=35, Resolution=224x2242022.0377.8