List your company

Action Recognition on Something-Something V1

70Top-1 AccInternVideo
45.40451.789558.17564.5605Jun 7, 2019Jan 24, 2020Sep 12, 2020May 2, 2021Dec 20, 2021Aug 9, 2022Mar 29, 2023

Evaluation results

MethodLinks
InternVideo2022.1270—
VideoMAE V2-gBackbone=ViT-g2023.0368.791.9
VideoMAE V2-HBackbone=ViT-H2023.0366.690.8
UniFormer2023.036187.6
VideoMAE2022.1260.9—
TDNBackbone=ResNet101, Frames=8+16, GFLOPS=1982020.1256.884.1
TDN2023.0356.884.1
RSANet-R50ENpre-trained=IN, #frame=8+16, FLOPS x clips=108 Gx22021.1156.182.8
RSANet-R50ENpre-trained=IN, #frame=8+16, FLOPS x clips=108 Gx12021.1155.582.6
TDNBackbone=ResNet101, Frames=16, GFLOPS=1322020.1255.383.3
GSMBackbone=Inception V3, Frames=fusion, GFLOPS=2682020.1255.2—
GSM2023.0355.2—
TDNBackbone=ResNet50, Frames=8+16, GFLOPS=1082020.1255.182.9
RNL TSMEnBackbone=R101 + R50, Frame x Clip x Crop=144=(8+16)x2x32020.0754.182.2
TDNBackbone=ResNet101, Frames=8, GFLOPS=662020.1254.181.9
RSANet-R50pre-trained=IN, #frame=16, FLOPS x clips=72 Gx12021.115481.1
TDNBackbone=ResNet50, Frames=16, GFLOPS=722020.1253.982.1
spatial-temporal discriminative filter banksBackbone=ResNet152, Pre-trained=ImageNet2019.0853.481.8
CorrNet-101Pretrain=Sports1M, Two stream=False2019.0653.3—
CorrNetBackbone=ResNet101, Frames=32x30, GFLOPS=224x30, Pre-trained=Sports1M2020.1253.3—
CorrNet2023.0353.3—
MARS+RGB+FlowPretrain=Kinetics, Two stream=True2019.0653—
RNL TSMEnBackbone=ResNet-50, Frame x Clip x Crop=144=(8+16)x2x32020.0752.781.5
TEINetBackbone=ResNet50, Frames=8+16, GFLOPS=992020.1252.5—
CT-Net-R50pre-trained=IN, #frame=16, FLOPS x clips=75 Gx12021.1152.580.9
TEINet2023.0352.5—
TDNBackbone=ResNet50, Frames=8, GFLOPS=362020.1252.380.6
TEApre-trained=IN, #frame=16, FLOPS x clips=70 Gx32021.1152.381.9
MSNet-TSM-R50pre-trained=IN, #frame=16, FLOPS x clips=67 Gx12021.1152.182.3
TEABackbone=ResNet50, Frames=16, GFLOPS=702020.1251.980.3
RSANet-R50pre-trained=IN, #frame=8, FLOPS x clips=36 Gx12021.1151.979.6
TEA2023.0351.980.3
MARS+RGBPretrain=Kinetics, Two stream=False2019.0651.7—
CorrNet-101Pretrain=None, Two stream=False, Note=Multi-clip sampling2019.0651.7—
CorrNetBackbone=ResNet101, Frames=32x30, GFLOPS=224x302020.1251.7—
SmallBigEnBackbone=ResNet-50, Frame x Clip x Crop=144=(8+16)x2x32020.0751.480.7
RNL TSMEnBackbone=ResNet-50, Frame x Clip x Crop=24=(8+16)x1x12020.0751.380.6
RNL TSMBackbone=ResNet-50, Frame x Clip x Crop=96=16x2x32020.075180.3
CorrNet-101Pretrain=None, Two stream=False2019.0650.9—
MSNet-R50Backbone=TSM-ResNet50, #3D=false, #Optical flow=false, #Frames=82021.0350.9—
CorrNet-R101#frame=32, FLOPS x clips=187 Gx102021.1150.9—
RNL TSMBackbone=ResNet-101, Frame x Clip x Crop=48=8x2x32020.0750.879.8
STMBackbone=ResNet50, Frames=16x30, GFLOPS=67x302020.1250.780.4
STM-R50pre-trained=IN, #frame=16, FLOPS x clips=67 Gx302021.1150.780.4
TANetBackbone=ResNet50, Frames=8+16, GFLOPS=992020.1250.679.3
TANet2023.0350.679.3
V4DBackbone=ResNet50, Frames=8x4, GFLOPS=167.62020.1250.4—
SmallBigNetBackbone=ResNet50, Frames=8+16, GFLOPS=1572020.1250.480.5
V4D2023.0350.4—
spatial-temporal discriminative filter banksBackbone=ResNet50, Pre-trained=ImageNet2019.0850.179.5
DFB-NetBackbone=3D ResNet50, #3D=true, #Optical flow=false, #Frames=162021.0350.1—
AFNet-TSM (RT=0.8)Dynamic=true, Backbone=AF-ResNet101, Frames=12, GFLOPS=48.92022.1150.1—
SmallBigBackbone=ResNet-50, Frame x Clip x Crop=96=16x2x32020.075079.8
AFNet-TSM (RT=0.8)Dynamic=true, Backbone=AF-ResNet50, Frames=12, GFLOPS=31.72022.1149.9—
AFNet-TSM (RT=0.4)Dynamic=true, Backbone=AF-ResNet101, Frames=12, GFLOPS=42.12022.1149.8—
TRGBackbone=Inception-V3, # Frames=32, # Params=28.3M, FLOPS=187.3G2019.0849.785.3
TSMEnBackbone=ResNet-50, Frame x Clip x Crop=24=(8+16)x1x12020.0749.778.5
TSMBackbone=ResNet50, Frames=8+16, GFLOPS=982020.1249.778.5
TSM2023.0349.778.5
TRGBackbone=ResNet-50, # Frames=32, # Params=28.4M, FLOPS=132.2G2019.0849.586.1
RNL TSMBackbone=ResNet-50, Frame x Clip x Crop=48=8x2x32020.0749.578.4
ECO-En Lite-RGB+FlowBackbone=BNIncep+3D Res18, #3D=true, #Optical flow=true, #Frames=92+922021.0349.5—
Ours-cumulativeBackbone=TSM-ResNet50, #3D=false, #Optical flow=false, #Frames=all, #Clusters=82021.0349.5—
RNL TSMBackbone=ResNet-50, Frame x Clip x Crop=16=16x1x12020.0749.4—
CorrNet-50Pretrain=None, Two stream=False2019.0649.3—
CorrNetBackbone=ResNet50, Frames=32x10, GFLOPS=115x102020.1249.3—
ir-CSN-152#frame=32, FLOPS x clips=97 Gx102021.1149.3—
STMBackbone=ResNet50, #3D=false, #Optical flow=false, #Frames=82021.0349.2—
AFNet-TSM (RT=0.4)Dynamic=true, Backbone=AF-ResNet50, Frames=12, GFLOPS=27.92022.1149—
TEADynamic=false, Backbone=ResNet50, Frames=8, GFLOPS=352022.1148.9—
GST-50Pretrain=ImageNet, Two stream=False2019.0648.6—
GSTBackbone=ResNet50, Frames=16, GFLOPS=592020.1248.677.9
TAMBackbone=bLResNet50, Frames=16x2, GFLOPS=47.72020.1248.478.8
SmallBigBackbone=ResNet-50, Frame x Clip x Crop=48=8x2x32020.0748.378.1
S3D-GPretrain=ImageNet, Two stream=True2019.0648.2—
S3D-GBackbone=Inception, #3D=true, #Optical flow=false, #Frames=642021.0348.2—
S3D-GBackbone=Inception, Frames=64, GFLOPS=71.382020.1248.278.7
TRGBackbone=ResNet-50, # Frames=16, # Params=28.4M, FLOPS=66.4G2019.0848.180.4
AdaFocus-TSMDynamic=true, Backbone=MobileNetV2+ResNet50, Frames=8+12, GFLOPS=33.72022.1148.1—
TSM + CGCBackbone=ResNet-50, Param=24.09M, Frame=162019.1047.8777.22
TrajectoryNetBackbone=ResNet18, Pre-trained=Kinetics-4002019.0847.8—
TrajectoryNetBackbone=S3DResNet-18, # Frames=162019.0847.8—
bLVNet-TAMDynamic=false, Backbone=bLResNet50, Frames=16, GFLOPS=35.12022.1147.8—
TRGBackbone=Inception, # Frames=32, # Params=14.1M, FLOPS=64.5G2019.0847.580.2
RNL TSMBackbone=ResNet-50, Frame x Clip x Crop=8=8x1x12020.0747.3—
TANetDynamic=false, Backbone=ResNet50, Frames=8, GFLOPS=332022.1147.3—
TRGBackbone=Inception-V3, # Frames=16, # Params=28.3M, FLOPS=94.6G2019.0847.278.9
TSMBackbone=ResNet-50, Frame x Clip x Crop=16=16x1x12020.0747.277.1
TSM-R50pre-trained=IN, #frame=16, FLOPS x clips=65 Gx12021.1147.277.1
SmallBigDynamic=false, Backbone=ResNet50, Frames=8, GFLOPS=522022.1147—
AdaFuse-TSMDynamic=true, Backbone=ResNet50, Frames=8, GFLOPS=31.52022.1146.8—
P3D + CGC 1D & 3DBackbone=ResNet-50, Param=25.62M, Frame=32 x 302019.1046.7376.04
Ours-slopeBackbone=TSM-ResNet50, #3D=false, #Optical flow=false, #Frames=all, #Clusters=82021.0346.7—
TSMBackbone=ResNet-50, Param=23.86M, Frame=162019.1046.6176.18
ECO liteBackbone=BN-Inception+ResNet18, Pre-trained=Kinetics-4002019.0846.4—
ECOenBackbone=Inception+3DResNet-18, # Frames=92, # Params=150.0M, FLOPS=267G2019.0846.4—
ECO En LiteBackbone=BNInc+Res18, Param=150M, Frame=922019.1046.4—
ECO-En LiteBackbone=BNIncep+3D Res18, #3D=true, #Optical flow=false, #Frames=922021.0346.4—
ECO_En LiteBackbone=BNIncep+R18, Frames=92, GFLOPS=2672020.1246.4—
P3D + CGC 3DBackbone=ResNet-50, Param=25.61M, Frame=32 x 302019.1046.3575.97