SOTA Video Classification on UCF101 (averaged over three splits) and PapersWithCode

98.7Accuracy

SCT-L

Updated 1mo ago

Evaluation Results

Method	Links
SCT-L 2021.08		98.7
SCT-M 2021.08		98.5
SCT-S 2021.08		98.3
SCT-S 2021.08		98
Two-Stream I3D [11] 2026.01		98
SCT-L 2021.08		97.7
LGD-3D 2021.08		97
R(2+1)D-RGB 2018.10		96.8
R(2+1)D 2021.08		96.8
S3D-G 2021.08		96.8
A2-Net 2018.10		96.4
OFF [14] 2026.01		96
I3D-RGB 2018.10		95.6
I3D 2021.08		95.4
ST-ResNet + IDT [33] 2026.01		94.6
ResNeXt 2021.08		94.5
Three-Stream TSN [20] 2026.01		94.2
Two-Stream TSN [20] 2026.01		94
Our (MViTv2-S based) 2026.01		93.54
Two-Stream CNNs [12] 2026.01		93.5
Two-Stream I3D [11] 2026.01		93.4
MViTv2-S finetune (our baseline) [5] 2026.01		92.11
LTC+iDT [32] 2026.01		91.7
TDD+iDT [31] 2026.01		91.5
Two-Stream C3D [13] 2026.01		91.4
MViTv1-B finetune (our baseline)[4] 2026.01		89.66
Two-Stream+LSTM [15] 2026.01		88.6
Two-Stream CNNs [18] 2026.01		88
Res3D 2018.10		85.8
VGG16 + Images 2015.12		83.5
VGG19 + Images 2015.12		83.4
VGG19 + Images 2015.12		83.3
Swin-S finetune (our baseline)[26] 2026.01		82.64
VGG16 + Images 2015.12		82.5
C3D 2018.10		82.3
VGG16 2015.12		77.9
VGG19 2015.12		77.8
spatial CNN 2015.12		73
slow fusion CNN 2015.12		65.4