List your company

Optical Flow on MPI Sintel Clean (test)

1.35AEELiteFlowNet+ft-arXiv
1.21122.14813.0854.0219Sep 20, 2017Apr 2, 2018Oct 14, 2018Apr 26, 2019Nov 7, 2019May 19, 2020Nov 30, 2020

Evaluation results

MethodLinks
LiteFlowNet+ft-arXivLearning=Supervised, Split=train2019.041.35———
FlowNet2+ftLearning=Supervised, Split=train2019.041.45———
LiteFlowNet+ft-CVPRLearning=Supervised, Split=train2019.041.64———
SelFlow+ftLearning=Supervised, Split=train2019.041.68———
PWC-Net+ft-axXivLearning=Supervised, Split=train2019.041.71———
ProFlow+ftLearning=Supervised, Split=train2019.041.78———
PWC-Net+ft-CVPRLearning=Supervised, Split=train2019.042.02———
MaskFlownetTime (s)=0.062020.032.52———
MRFlowTime (s)=4802018.092.53———
MRFlowTime (s)=4802018.122.53———
MR-FlowTime (s)=4802020.032.53———
ProFlow ROBEvaluation Context=Robust Vision Challenge2018.112.7091.01316.549—
FullFlowTime (s)=2402018.092.71———
FullFlowTime (s)=2402018.122.71———
MaskFlownet-STime (s)=0.032020.032.77———
VCNTime (s)=0.182020.032.81———
ProFlow+ftLearning=Supervised, Split=test2019.042.82———
SelFlowLearning=Unsupervised, Split=train2019.042.88———
DDFlow+ftLearning=Unsupervised, Split=train2019.042.92———
LiteFlowNet32020.072.99———
LiteFlowNet3-S2020.073.03———
SMURFheavier backbone=true, multi-frame=true2020.113.15———
SpyNet+ftLearning=Supervised, Split=train2019.043.17———
VCN-small2020.073.26———
ContinualFlow+ftLearning=Supervised, Split=test2019.043.34———
ContinualFlow_ROBEvaluation Context=Robust Vision Challenge2018.113.3411.75216.292—
MFF+ftLearning=Supervised, Split=test2019.043.42———
PWC-Net+Time (s)=0.03, fine-tuned=true, protocol=training protocol II2018.093.45———
PWC-Net+ft-axXivLearning=Supervised, Split=test2019.043.45———
LiteFlowNet2Time (s)=0.042020.033.45———
PWC-Net+Time (s)=0.032020.033.45———
PWC-Net+2020.073.45———
LiteFlowNet22020.073.48———
DCFlowTime (s)=8.602018.093.54———
DCFlowTime (s)=8.62018.123.54———
DCFlow+ftLearning=Supervised, Split=test2019.043.54———
CPM-flowTime (s)=4.302018.093.56———
CPM-flowTime (s)=4.32018.123.56———
SENSEmulti-frame=true2020.073.6———
AugFNG_ROBEvaluation Context=Robust Vision Challenge, Status=anonymous2018.113.6061.60319.939—
FlowNetS+ftLearning=Supervised, Split=train2019.043.66———
SelFlow+ftLearning=Supervised, Split=test2019.043.74———
SelFlowTime (s)=0.092020.033.74———
SelFlowLearning Paradigm=Supervised, Fine-tuned=true, Multi-frame=true, Number of Parameters=4.79 M2020.033.74———
FlowFieldsCategory=Conventional2018.053.75———
FlowFieldsTime (s)=28.02018.093.75———
FlowFieldsTime (s)=282018.123.75———
SelFlowmulti-frame=true2020.073.75———
Bailer et al.Category=Hybrid2018.053.78———
FlowFieldsCNNTime (s)=23.02018.093.78———
FlowFieldsCNNTime (s)=232018.123.78———
FlowNetC+ftLearning=Supervised, Split=train2019.043.78———
FlowFieldsCNN+ftLearning=Supervised, Split=test2019.043.78———
IRR-PWCLearning Paradigm=Supervised, Fine-tuned=true, Number of Parameters=6.36 M2020.033.84———
IRR-PWC2020.073.84———
Deep DiscreteFlowCategory=Hybrid2018.053.86———
PWC-Net-ftTime (s)=0.03, fine-tuned=true2018.093.86———
PWC-NetTime (s)=0.032020.033.86———
MultiFrameOccFlow-Soft+ftLearning=Unsupervised, Split=train2019.043.89———
PWC-Net_ROBEvaluation Context=Robust Vision Challenge2018.113.8971.72621.637—
PWC-Net_ROBTime (s)=0.03, fine-tuned=true, data=Sintel, KITTI 2015, HD1K2018.093.9———
FF++_ROBEvaluation Context=Robust Vision Challenge2018.113.9531.14826.836—
FlowNet2Category=Heavyweight CNN2018.053.96———
FlowNet2Time (s)=0.122018.093.96———
FlowNet2Fine-tuned=false, Time (s)=0.122018.123.96———
FlowNet2Time (s)=0.122020.033.96———
CoL-AMFlowheavier backbone=true2020.113.96———
OccAwareFlow+ftLearning=Unsupervised, Split=train2019.044.03———
LiteFlowNetCategory=Lightweight CNN2018.054.04———
EpicFlow2017.094.12———
EpicFlowCategory=Conventional2018.054.12———
EpicFlowTime (s)=15.02018.094.12———
EpicFlowTime (s)=152018.124.12———
FlowNet2-ft-sintelCategory=Heavyweight CNN, fine-tuned=Sintel2018.054.16———
FlowNet2-ftTime (s)=0.12, fine-tuned=true2018.094.16———
FlowNet2-ftFine-tuned=true, Time (s)=0.122018.124.16———
FlowNet2+ftLearning=Supervised, Split=test2019.044.16———
FlowNet22020.074.16———
Devon2020.074.34———
FlowNet32020.074.35———
PWC-Net-ft-finalTime (s)=0.03, fine-tuned=true, protocol=more weight to the final pass2018.094.39———
PWC-Net-ftFine-tuned=true, Time (s)=0.032018.124.39———
PWC-Net+ft-CVPRLearning=Supervised, Split=test2019.044.39———
PWC-NetLearning Paradigm=Supervised, Fine-tuned=true, Number of Parameters=8.75 M2020.034.39———
PWC-Net2020.074.39———
ARFlow-MVLearning Paradigm=Unsupervised, Multi-frame=true, Number of Parameters=2.37 M2020.034.49———
ARFlow-MVmulti-frame=true2020.114.49———
LiteFlowNet-ftCategory=Lightweight CNN, fine-tuned=true2018.054.54———
LiteFlowNet-arXivTime (s)=0.092018.094.54———
LiteFlowNet+ft-arXivLearning=Supervised, Split=test2019.044.54———
LiteFlowNetTime (s)=0.092020.034.54———
LiteFlowNet2020.074.54———
UPFlowheavier backbone=true2020.114.68———
UnrolledCost2020.114.69———
ARFlowLearning Paradigm=Unsupervised, Number of Parameters=2.24 M2020.034.78———
ARFlow2020.114.78———
HD3F-ftFine-tuned=true, Time (s)=0.082018.124.79———
HD³multi-frame=true2020.074.79———
LFNet_ROBEvaluation Context=Robust Vision Challenge, Status=anonymous2018.114.8152.33325.065—
LiteFlowNetXCategory=Lightweight CNN2018.054.82———