SOTAVerified

Optical Flow Estimation

Optical Flow Estimation is a computer vision task that involves computing the motion of objects in an image or a video sequence. The goal of optical flow estimation is to determine the movement of pixels or features in the image, which can be used for various applications such as object tracking, motion analysis, and video compression.

Approaches for optical flow estimation include correlation-based, block-matching, feature tracking, energy-based, and more recently gradient-based.

Further readings:

Definition source: Devon: Deformable Volume Network for Learning Optical Flow

Image credit: Optical Flow Estimation

Papers

Showing 401450 of 2184 papers

TitleStatusHype
Self-Supervised Learning of Audio-Visual Objects from VideoCode1
PAN: Towards Fast Action Recognition via Learning Persistence of AppearanceCode1
Memory-augmented Dense Predictive Coding for Video Representation LearningCode1
Entropy Minimisation Framework for Event-based Vision Model EstimationCode1
Unsupervised Learning of Particle Image VelocimetryCode1
Robust Ego and Object 6-DoF Motion Estimation and TrackingCode1
MuCAN: Multi-Correspondence Aggregation Network for Video Super-ResolutionCode1
LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow EstimationCode1
STaRFlow: A SpatioTemporal Recurrent Cell for Lightweight Multi-Frame Optical Flow EstimationCode1
Learning Geocentric Object Pose in Oblique Monocular ImagesCode1
Creating Artificial Modalities to Solve RGB LivenessCode1
Motion Representation Using Residual Frames with 3D CNNCode1
Video Semantic Segmentation with Distortion-Aware Feature CorrectionCode1
Multiple Video Frame Interpolation via Enhanced Deformable Separable ConvolutionCode1
What Matters in Unsupervised Optical FlowCode1
End-to-end Learning for Inter-Vehicle Distance and Relative Velocity Estimation in ADAS with a Monocular CameraCode1
Robust Reference-Based Super-Resolution With Similarity-Aware Deformable ConvolutionCode1
High-quality Panorama Stitching based on Asymmetric Bidirectional Optical FlowCode1
TDAN: Temporally-Deformable Alignment Network for Video Super-ResolutionCode1
Upgrading Optical Flow to 3D Scene Flow Through Optical ExpansionCode1
FeatureFlow: Robust Video Interpolation via Structure-to-Texture GenerationCode1
Video Frame Interpolation via Residue RefinementCode1
Rolling-Unrolling LSTMs for Action Anticipation from First-Person VideoCode1
IPN Hand: A Video Dataset and Benchmark for Real-Time Continuous Hand Gesture RecognitionCode1
Motion and Region Aware Adversarial Learning for Fall Detection with Thermal ImagingCode1
A Transductive Approach for Video Object SegmentationCode1
Self-Supervised Monocular Scene Flow EstimationCode1
Cascaded Deep Video Deblurring Using Temporal Sharpness PriorCode1
Flow2Stereo: Effective Self-Supervised Learning of Optical Flow and Stereo MatchingCode1
RANSAC-Flow: generic two-stage image alignmentCode1
Conquering the cnn over-parameterization dilemma: A volterra filtering approach for action recognitionCode1
Channel Attention Is All You Need for Video Frame InterpolationCode1
Towards Better Generalization: Joint Depth-Pose Learning without PoseNetCode1
Distilled Semantics for Comprehensive Scene Understanding from VideosCode1
Probabilistic Pixel-Adaptive Refinement NetworksCode1
Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow EstimationCode1
RAFT: Recurrent All-Pairs Field Transforms for Optical FlowCode1
MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion MaskCode1
Spike-FlowNet: Event-based Optical Flow Estimation with Energy-Efficient Hybrid Neural NetworksCode1
Belief Propagation Reloaded: Learning BP-Layers for Labeling ProblemsCode1
AirSim Drone Racing LabCode1
Softmax Splatting for Video Frame InterpolationCode1
Occlusion Aware Unsupervised Learning of Optical Flow From VideoCode1
Unsupervised Learning of Depth, Optical Flow and Pose with Occlusion from 3D GeometryCode1
Extremely Dense Point Correspondences using a Learned Feature DescriptorCode1
Deep Slow Motion Video Reconstruction with Hybrid Imaging SystemCode1
Efficient Semantic Video Segmentation with Per-frame InferenceCode1
ScopeFlow: Dynamic Scene Scoping for Optical FlowCode1
Semantic Flow for Fast and Accurate Scene ParsingCode1
Over-the-Air Adversarial Flickering Attacks against Video Recognition NetworksCode1
Show:102550
← PrevPage 9 of 44Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SpynetAverage End-Point Error6.64Unverified
2FastFlowNet-ftAverage End-Point Error4.89Unverified
3UnrolledCostAverage End-Point Error4.69Unverified
4LiteFlowNet-ftAverage End-Point Error4.54Unverified
5FlowNet2Average End-Point Error3.96Unverified
6IRR-PWCAverage End-Point Error3.84Unverified
7SelFlowAverage End-Point Error3.74Unverified
8FDFlowNet-ftAverage End-Point Error3.71Unverified
9ScopeFlowAverage End-Point Error3.59Unverified
10LiteFlowNet2-ftAverage End-Point Error3.48Unverified
#ModelMetricClaimedVerifiedStatus
1SpynetAverage End-Point Error8.36Unverified
2FastFlowNet-ftAverage End-Point Error6.08Unverified
3UnrolledCostAverage End-Point Error5.8Unverified
4MR-FlowAverage End-Point Error5.38Unverified
5LiteFlowNet-ftAverage End-Point Error5.38Unverified
6FDFlowNet-ftAverage End-Point Error5.11Unverified
7LiteFlowNet2-ftAverage End-Point Error4.69Unverified
8IRR-PWCAverage End-Point Error4.58Unverified
9LiteFlowNet3-SAverage End-Point Error4.53Unverified
10ContinualFlow + ftAverage End-Point Error4.52Unverified
#ModelMetricClaimedVerifiedStatus
1PWC-NetF1-all33.7Unverified
2FastFlowNetF1-all33.1Unverified
3FlowNet2F1-all30Unverified
4VCNF1-all25.1Unverified
5HD3F1-all24Unverified
6MaskFlowNetF1-all23.1Unverified
7SCVF1-all19.3Unverified
8RAPIDFlowF1-all17.7Unverified
9CRAFTF1-all17.5Unverified
10RAFTF1-all17.4Unverified
#ModelMetricClaimedVerifiedStatus
1FastFlowNet-ftFl-all11.22Unverified
2UnrolledCostFl-all10.81Unverified
3LiteFlowNet-ftFl-all9.38Unverified
4SelFlowFl-all8.42Unverified
5IRR-PWCFl-all7.65Unverified
6LiteFlowNet2-ftFl-all7.62Unverified
7LiteFlowNet3Fl-all7.34Unverified
8LiteFlowNet3-SFl-all7.22Unverified
9MaskFlownet-SFl-all6.81Unverified
10RAPIDFlowFl-all6.12Unverified
#ModelMetricClaimedVerifiedStatus
1FastFlowNet-ftAverage End-Point Error1.8Unverified
2LiteFlowNet-ftAverage End-Point Error1.6Unverified
3IRR-PWCAverage End-Point Error1.6Unverified
4SelFlowAverage End-Point Error1.5Unverified
5FDFlowNet-ftAverage End-Point Error1.5Unverified
6PWC-Net + ft - axXivAverage End-Point Error1.5Unverified
7LiteFlowNet2-ftAverage End-Point Error1.4Unverified
8LiteFlowNet3-SAverage End-Point Error1.3Unverified
9LiteFlowNet3Average End-Point Error1.3Unverified
10MaskFlownetAverage End-Point Error1.1Unverified
#ModelMetricClaimedVerifiedStatus
1PWCNet1px total82.27Unverified
2SPyNet1px total29.96Unverified
3GMFlow1px total10.36Unverified
4GMA1px total7.07Unverified
5RAFT1px total6.79Unverified
6FlowNet21px total6.71Unverified
7FlowFormer1px total6.51Unverified
8MS-RAFT+1px total5.72Unverified
9RPKNet1px total4.81Unverified
10DPFlow1px total3.44Unverified
#ModelMetricClaimedVerifiedStatus
1UFlowAverage End-Point Error5.21Unverified
2MDFlow-FastAverage End-Point Error4.73Unverified
3UpFlowAverage End-Point Error4.68Unverified
4ARFlow-MVAverage End-Point Error4.49Unverified
5MDFlowAverage End-Point Error4.16Unverified
#ModelMetricClaimedVerifiedStatus
1UFlowAverage End-Point Error6.5Unverified
2MDFlow-FastAverage End-Point Error5.99Unverified
3ARFlow-MVAverage End-Point Error5.67Unverified
4MDFlowAverage End-Point Error5.46Unverified
5UpFlowAverage End-Point Error5.32Unverified
#ModelMetricClaimedVerifiedStatus
1ARFlow-MVFl-all11.79Unverified
2MDFlow-FastFl-all11.43Unverified
3UpFlowFl-all9.38Unverified
4MDFlowFl-all8.91Unverified
#ModelMetricClaimedVerifiedStatus
1ARFlow-MVAverage End-Point Error1.5Unverified
2UpFlowAverage End-Point Error1.4Unverified