Optical Flow Estimation On Sintel Final

评估指标

Average End-Point Error

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
Spynet	8.36	Optical Flow Estimation using a Spatial Pyramid Network
FastFlowNet-ft	6.08	FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation
UnrolledCost	5.8	Cost Function Unrolling in Unsupervised Optical Flow
MR-Flow	5.38	Optical Flow in Mostly Rigid Scenes	-
LiteFlowNet-ft	5.38	LiteFlowNet: A Lightweight Convolutional Neural Network for Optical Flow Estimation
FDFlowNet-ft	5.11	FDFlowNet: Fast Optical Flow Estimation using a Deep Lightweight Network	-
LiteFlowNet2-ft	4.69	A Lightweight Optical Flow CNN - Revisiting Data Fidelity and Regularization
IRR-PWC	4.579	Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimation
LiteFlowNet3-S	4.53	LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow Estimation
ContinualFlow + ft	4.52	Continual Occlusions and Optical Flow Estimation	-
LiteFlowNet3	4.45	LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow Estimation
VCN	4.40	Volumetric Correspondence Networks for Optical Flow	-
MaskFlownet-S	4.38	MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Mask
SelFlow	4.26	SelFlow: Self-Supervised Learning of Optical Flow
MaskFlownet	4.17	MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Mask
ScopeFlow	4.098	ScopeFlow: Dynamic Scene Scoping for Optical Flow
RAPIDFlow	3.56	RAPIDFlow: Recurrent Adaptable Pyramids with Iterative Decoding for Efficient Optical Flow Estimation	-
DEQ-Flow-H	2.886	Deep Equilibrium Optical Flow Estimation
RAFT (warm-start)	2.855	RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
RPKNet	2.657	Recurrent Partial Kernel Network for Efficient Optical Flow Estimation	-

0 of 27 row(s) selected.