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

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1951–2000 of 2184 papers

TitleStatusHype
Accurate Optical Flow via Direct Cost Volume Processing—0
Learn to Model Motion from Blurry Footages—0
Joint Semantic and Motion Segmentation for dynamic scenes using Deep Convolutional Networks—0
Robust Optical Flow Estimation in Rainy Scenes—0
A Nuclear-norm Model for Multi-Frame Super-Resolution Reconstruction from Video Clips—0
Video Acceleration MagnificationCode0
Predictive-Corrective Networks for Action Detection—0
Pyramidal Gradient Matching for Optical Flow Estimation—0
Hidden Two-Stream Convolutional Networks for Action RecognitionCode1
Google Map Aided Visual Navigation for UAVs in GPS-denied Environment—0
Saliency-guided video classification via adaptively weighted learning—0
Video Frame Interpolation via Adaptive ConvolutionCode0
Predicting Deeper into the Future of Semantic SegmentationCode0
Spatio-Temporal Facial Expression Recognition Using Convolutional Neural Networks and Conditional Random Fields—0
Learning Rank Reduced Interpolation with Principal Component Analysis—0
Flow Fields: Dense Correspondence Fields for Highly Accurate Large Displacement Optical Flow Estimation—0
Instance Flow Based Online Multiple Object Tracking—0
Optical Flow-based 3D Human Motion Estimation from Monocular Video—0
Boundary Flow: A Siamese Network that Predicts Boundary Motion without Training on Motion—0
An Analysis of Parallelized Motion Masking Using Dual-Mode Single Gaussian ModelsCode0
One-Step Time-Dependent Future Video Frame Prediction with a Convolutional Encoder-Decoder Neural Network—0
Guided Optical Flow Learning—0
Video Frame Synthesis using Deep Voxel FlowCode0
Detailed Surface Geometry and Albedo Recovery from RGB-D Video Under Natural Illumination—0
Video Salient Object Detection via Fully Convolutional Networks—0
Vertical Landing for Micro Air Vehicles using Event-Based Optical Flow—0
Real-Time Optical flow-based Video Stabilization for Unmanned Aerial Vehicles—0
Ordered Pooling of Optical Flow Sequences for Action Recognition—0
Distinguishing Posed and Spontaneous Smiles by Facial Dynamics—0
Semantic Video Segmentation by Gated Recurrent Flow Propagation—0
Learning Motion Patterns in Videos—0
Deep Motion Features for Visual Tracking—0
Objective Micro-Facial Movement Detection Using FACS-Based Regions and Baseline Evaluation—0
SceneNet RGB-D: 5M Photorealistic Images of Synthetic Indoor Trajectories with Ground TruthCode0
A Video-Based Method for Objectively Rating Ataxia—0
Hybrid Learning of Optical Flow and Next Frame Prediction to Boost Optical Flow in the Wild—0
ActionFlowNet: Learning Motion Representation for Action Recognition—0
A Maximum A Posteriori Estimation Framework for Robust High Dynamic Range Video Synthesis—0
DeMoN: Depth and Motion Network for Learning Monocular StereoCode0
FlowNet 2.0: Evolution of Optical Flow Estimation with Deep NetworksCode0
Multi-way Particle Swarm Fusion—0
Action Recognition with Dynamic Image NetworksCode0
Deep Alternative Neural Network: Exploring Contexts as Early as Possible for Action Recognition—0
Proximal Deep Structured Models—0
Surveillance Video Parsing with Single Frame Supervision—0
InterpoNet, A brain inspired neural network for optical flow dense interpolationCode0
AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks for Human Action Recognition in Videos—0
AutoScaler: Scale-Attention Networks for Visual Correspondence—0
Optical Flow Requires Multiple Strategies (but only one network)Code0
Estimating motion with principal component regression strategies—0
Show:102550
← PrevPage 40 of 44Next →
All datasetsSintel-cleanSintel-finalKITTI 2015 (train)KITTI 2015KITTI 2012SpringSintel Clean unsupervisedSintel Final unsupervisedKITTI 2015 unsupervisedKITTI 2012 unsupervised

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SpynetAverage End-Point Error6.64—Unverified
2FastFlowNet-ftAverage End-Point Error4.89—Unverified
3UnrolledCostAverage End-Point Error4.69—Unverified
4LiteFlowNet-ftAverage End-Point Error4.54—Unverified
5FlowNet2Average End-Point Error3.96—Unverified
6IRR-PWCAverage End-Point Error3.84—Unverified
7SelFlowAverage End-Point Error3.74—Unverified
8FDFlowNet-ftAverage End-Point Error3.71—Unverified
9ScopeFlowAverage End-Point Error3.59—Unverified
10LiteFlowNet2-ftAverage End-Point Error3.48—Unverified
#ModelMetricClaimedVerifiedStatus
1SpynetAverage End-Point Error8.36—Unverified
2FastFlowNet-ftAverage End-Point Error6.08—Unverified
3UnrolledCostAverage End-Point Error5.8—Unverified
4LiteFlowNet-ftAverage End-Point Error5.38—Unverified
5MR-FlowAverage End-Point Error5.38—Unverified
6FDFlowNet-ftAverage End-Point Error5.11—Unverified
7LiteFlowNet2-ftAverage End-Point Error4.69—Unverified
8IRR-PWCAverage End-Point Error4.58—Unverified
9LiteFlowNet3-SAverage End-Point Error4.53—Unverified
10ContinualFlow + ftAverage End-Point Error4.52—Unverified
#ModelMetricClaimedVerifiedStatus
1PWC-NetF1-all33.7—Unverified
2FastFlowNetF1-all33.1—Unverified
3FlowNet2F1-all30—Unverified
4VCNF1-all25.1—Unverified
5HD3F1-all24—Unverified
6MaskFlowNetF1-all23.1—Unverified
7SCVF1-all19.3—Unverified
8RAPIDFlowF1-all17.7—Unverified
9CRAFTF1-all17.5—Unverified
10RAFTF1-all17.4—Unverified
#ModelMetricClaimedVerifiedStatus
1FastFlowNet-ftFl-all11.22—Unverified
2UnrolledCostFl-all10.81—Unverified
3LiteFlowNet-ftFl-all9.38—Unverified
4SelFlowFl-all8.42—Unverified
5IRR-PWCFl-all7.65—Unverified
6LiteFlowNet2-ftFl-all7.62—Unverified
7LiteFlowNet3Fl-all7.34—Unverified
8LiteFlowNet3-SFl-all7.22—Unverified
9MaskFlownet-SFl-all6.81—Unverified
10RAPIDFlowFl-all6.12—Unverified
#ModelMetricClaimedVerifiedStatus
1FastFlowNet-ftAverage End-Point Error1.8—Unverified
2IRR-PWCAverage End-Point Error1.6—Unverified
3LiteFlowNet-ftAverage End-Point Error1.6—Unverified
4PWC-Net + ft - axXivAverage End-Point Error1.5—Unverified
5FDFlowNet-ftAverage End-Point Error1.5—Unverified
6SelFlowAverage End-Point Error1.5—Unverified
7LiteFlowNet2-ftAverage End-Point Error1.4—Unverified
8LiteFlowNet3Average End-Point Error1.3—Unverified
9LiteFlowNet3-SAverage End-Point Error1.3—Unverified
10MaskFlownet-SAverage End-Point Error1.1—Unverified
#ModelMetricClaimedVerifiedStatus
1PWCNet1px total82.27—Unverified
2SPyNet1px total29.96—Unverified
3GMFlow1px total10.36—Unverified
4GMA1px total7.07—Unverified
5RAFT1px total6.79—Unverified
6FlowNet21px total6.71—Unverified
7FlowFormer1px total6.51—Unverified
8MS-RAFT+1px total5.72—Unverified
9RPKNet1px total4.81—Unverified
10DPFlow1px total3.44—Unverified
#ModelMetricClaimedVerifiedStatus
1UFlowAverage End-Point Error5.21—Unverified
2MDFlow-FastAverage End-Point Error4.73—Unverified
3UpFlowAverage End-Point Error4.68—Unverified
4ARFlow-MVAverage End-Point Error4.49—Unverified
5MDFlowAverage End-Point Error4.16—Unverified
#ModelMetricClaimedVerifiedStatus
1UFlowAverage End-Point Error6.5—Unverified
2MDFlow-FastAverage End-Point Error5.99—Unverified
3ARFlow-MVAverage End-Point Error5.67—Unverified
4MDFlowAverage End-Point Error5.46—Unverified
5UpFlowAverage End-Point Error5.32—Unverified
#ModelMetricClaimedVerifiedStatus
1ARFlow-MVFl-all11.79—Unverified
2MDFlow-FastFl-all11.43—Unverified
3UpFlowFl-all9.38—Unverified
4MDFlowFl-all8.91—Unverified
#ModelMetricClaimedVerifiedStatus
1ARFlow-MVAverage End-Point Error1.5—Unverified
2UpFlowAverage End-Point Error1.4—Unverified