List your company
ResearchFormerly WizWand

Optical Flow on MVSEC 1.0 (indoor_flying1)

0.11EPEEDCFlow (+in2)
-0.0581.0762.213.344Apr 17, 2019Jun 8, 2020Aug 1, 2021Sep 24, 2022Nov 17, 2023Jan 9, 2025Mar 4, 2026
Updated 8mo ago

Evaluation results

MethodLinks
EDCFlow (+in2)dt=1, Input=E, Learning paradigm=SL, training_inclusion=flying22025.060.112
ECN*dt=1, training_split=80%, evaluation_split=20%2020.030.210.01
ScaleEventBackbone=ViT-S/16, Protocol=Event Pretraining + Fully-Supervised2026.030.290.03
STPBackbone=Swin-T/7, Protocol=Event Pretraining + Fully-Supervised2026.030.310.03
ECDDPBackbone=Swin-T/7, Protocol=Event Pretraining + Fully-Supervised2026.030.360.04
EV-MGRFlowNetdt=1, Input=E, Learning paradigm=USL2025.060.4117
MultiCM (Burgers')dt=1, Input=E, Learning paradigm=MB2025.060.4210
EDCFlow (+in3)dt=1, Input=E, Learning paradigm=SL, training_inclusion=flying32025.060.421
ECDDPBackbone=ViT-S/16, Protocol=Event Pretraining + Fully-Supervised2026.030.510.11
Brebion et al.dt=1, Input=E, Learning paradigm=MB2025.060.5210
Oursdt=12020.030.561
STE-FlowNetdt=1, Input=E, Learning paradigm=SSL2025.060.5710
Zhu et al.dt=1, Input=E, Learning paradigm=USL2025.060.580
EVFlow+dt=12020.030.580
STPBackbone=ViT-S/16, Protocol=Event Pretraining + Fully-Supervised2026.030.580.05
ECDPBackbone=ResNet-50, Protocol=Event Pretraining + Fully-Supervised2026.030.60.35
ECDPBackbone=ViT-S/16, Protocol=Event Pretraining + Fully-Supervised2026.030.610.05
DCEIFlowdt=1, Input=E+I1, Learning paradigm=SL2025.060.75155
DCEFlowProtocol=RGB initialization + Fully-Supervised2026.030.750.6
Voxel GridMeasurement=polarity, Kernel=trilinear2019.040.961.47
ESTMeasurement=time stamps, Kernel=exponential2019.040.961.27
Gehrigdt=12020.030.960.91
EDCFlowdt=1, Input=E, Learning paradigm=SL2025.060.97219
ESTMeasurement=time stamps, Kernel=learnt2019.040.970.91
Voxel GridMeasurement=time stamps, Kernel=trilinear2019.040.981.2
ESTMeasurement=time stamps, Kernel=trilinear2019.0411.35
EV-FlowNetdt=1, Input=E, Learning paradigm=SSL2025.061.03220
EV-FlowNetMeasurement=count, Kernel=trilinear2019.041.032.2
ESTMeasurement=time stamps, Kernel=alpha2019.041.031.34
EVFlowdt=12020.031.032.2
IDNet-4dt=1, Input=E, Learning paradigm=SL, re-implemented=true2025.061.04186
TMAdt=1, Input=E, Learning paradigm=SL2025.061.06363
E-RAFTdt=1, Input=E, Learning paradigm=SL2025.061.1572
Zerosdt=12020.031.11
IDNet-8dt=1, Input=E, Learning paradigm=SL, re-implemented=true2025.061.13294
Event FrameMeasurement=time stamps, Kernel=trilinear2019.041.172.44
Two-Channel ImageMeasurement=time stamps, Kernel=trilinear2019.041.171.5
Two-Channel ImageMeasurement=count, Kernel=trilinear2019.041.214.49
ESTBackbone=ResNet-18, Protocol=RGB initialization + Fully-Supervised2026.031.245.09
EDCFlow (+in2)dt=4, Input=E, Learning paradigm=SL, training_inclusion=flying22025.061.28524
EDCFlow (+in3)dt=4, Input=E, Learning paradigm=SL, training_inclusion=flying32025.061.3556
EV-MGRFlowNetdt=4, Input=E, Learning paradigm=USL2025.061.5867
MultiCM (Burgers')dt=4, Input=E, Learning paradigm=MB2025.061.691,295
STE-FlowNetdt=4, Input=E, Learning paradigm=SSL2025.061.771,470
DCEIFlowdt=4, Input=E+I1, Learning paradigm=SL2025.062.082,147
Zhu et al.dt=4, Input=E, Learning paradigm=USL2025.062.182,420
EV-FlowNetdt=4, Input=E, Learning paradigm=SSL2025.062.252,470
TMAdt=4, Input=E, Learning paradigm=SL2025.062.432,991
EDCFlowdt=4, Input=E, Learning paradigm=SL2025.062.592,816
E-RAFTdt=4, Input=E, Learning paradigm=SL2025.062.814,025
IDNet-4dt=4, Input=E, Learning paradigm=SL, re-implemented=true2025.063.34,272
IDNet-8dt=4, Input=E, Learning paradigm=SL, re-implemented=true2025.064.315,895