HyperAI超神经
首页资讯最新论文教程数据集百科SOTALLM 模型天梯GPU 天梯顶会
全站搜索
关于
HyperAI超神经

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

模型名称
Average End-Point Error
Paper TitleRepository
Spynet8.36Optical Flow Estimation using a Spatial Pyramid Network
DEQ-Flow-H2.886Deep Equilibrium Optical Flow Estimation
SelFlow4.26SelFlow: Self-Supervised Learning of Optical Flow
FastFlowNet-ft6.08FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation
FDFlowNet-ft5.11FDFlowNet: Fast Optical Flow Estimation using a Deep Lightweight Network-
RPKNet2.657Recurrent Partial Kernel Network for Efficient Optical Flow Estimation
MR-Flow5.38Optical Flow in Mostly Rigid Scenes-
MaskFlownet4.17MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Mask
GMFlow2.37Unifying Flow, Stereo and Depth Estimation
CroCo-Flow2.436CroCo v2: Improved Cross-view Completion Pre-training for Stereo Matching and Optical Flow
VCN4.40Volumetric Correspondence Networks for Optical Flow
IRR-PWC4.579Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimation
Perceiver IO2.42Perceiver IO: A General Architecture for Structured Inputs & Outputs
Ef-RAFT2.60Rethinking RAFT for Efficient Optical Flow
UnrolledCost5.8Cost Function Unrolling in Unsupervised Optical Flow
GMFlowNet2.648Global Matching with Overlapping Attention for Optical Flow Estimation
ScopeFlow4.098ScopeFlow: Dynamic Scene Scoping for Optical Flow
GMA2.470Learning to Estimate Hidden Motions with Global Motion Aggregation
ContinualFlow + ft4.52Continual Occlusions and Optical Flow Estimation-
LiteFlowNet2-ft4.69A Lightweight Optical Flow CNN -- Revisiting Data Fidelity and Regularization
0 of 27 row(s) selected.