SOTAVerified

Motion Segmentation

Motion Segmentation is an essential task in many applications in Computer Vision and Robotics, such as surveillance, action recognition and scene understanding. The classic way to state the problem is the following: given a set of feature points that are tracked through a sequence of images, the goal is to cluster those trajectories according to the different motions they belong to. It is assumed that the scene contains multiple objects that are moving rigidly and independently in 3D-space.

Source: Robust Motion Segmentation from Pairwise Matches

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 212 papers

TitleStatusHype
Moving Object Segmentation: All You Need Is SAM (and Flow)Code3
RaTrack: Moving Object Detection and Tracking with 4D Radar Point CloudCode2
DytanVO: Joint Refinement of Visual Odometry and Motion Segmentation in Dynamic EnvironmentsCode2
ParticleSfM: Exploiting Dense Point Trajectories for Localizing Moving Cameras in the WildCode2
FreeGave: 3D Physics Learning from Dynamic Videos by Gaussian VelocityCode1
MOD-UV: Learning Mobile Object Detectors from Unlabeled VideosCode1
Motion2Language, unsupervised learning of synchronized semantic motion segmentationCode1
Bootstrapping Objectness from Videos by Relaxed Common Fate and Visual GroupingCode1
Hidden Gems: 4D Radar Scene Flow Learning Using Cross-Modal SupervisionCode1
Segmenting Moving Objects via an Object-Centric Layered RepresentationCode1
Discovering Objects that Can MoveCode1
HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object InteractionCode1
Self-Supervised Scene Flow Estimation with 4-D Automotive RadarCode1
EM-driven unsupervised learning for efficient motion segmentationCode1
Formulating Event-based Image Reconstruction as a Linear Inverse Problem with Deep Regularization using Optical FlowCode1
Monocular Arbitrary Moving Object Discovery and SegmentationCode1
Event-based Motion Segmentation by Cascaded Two-Level Multi-Model FittingCode1
Local Frequency Domain Transformer Networks for Video PredictionCode1
Video Class Agnostic Segmentation Benchmark for Autonomous DrivingCode1
SSTVOS: Sparse Spatiotemporal Transformers for Video Object SegmentationCode1
MultiBodySync: Multi-Body Segmentation and Motion Estimation via 3D Scan SynchronizationCode1
Learning to Segment Rigid Motions from Two FramesCode1
Event-based Motion Segmentation with Spatio-Temporal Graph CutsCode1
0-MMS: Zero-Shot Multi-Motion Segmentation With A Monocular Event CameraCode1
Understanding Dynamic Scenes using Graph Convolution NetworksCode1
Motion Segmentation using Frequency Domain Transformer NetworksCode1
Distilled Semantics for Comprehensive Scene Understanding from VideosCode1
Progressive-X: Efficient, Anytime, Multi-Model Fitting AlgorithmCode1
UnOS: Unified Unsupervised Optical-Flow and Stereo-Depth Estimation by Watching VideosCode1
Event-Based Motion Segmentation by Motion CompensationCode1
FlowNet3D: Learning Scene Flow in 3D Point CloudsCode1
Sparse Subspace Clustering: Algorithm, Theory, and ApplicationsCode1
Channel-wise Motion Features for Efficient Motion Segmentation—0
Temporal Rate Reduction Clustering for Human Motion Segmentation—0
KDMOS:Knowledge Distillation for Motion SegmentationCode0
EV-LayerSegNet: Self-supervised Motion Segmentation using Event Cameras—0
Event-based Egocentric Human Pose Estimation in Dynamic Environment—0
Iterative Event-based Motion Segmentation by Variational Contrast MaximizationCode0
Dynamic Point Maps: A Versatile Representation for Dynamic 3D Reconstruction—0
BEVMOSNet: Multimodal Fusion for BEV Moving Object Segmentation—0
Wandering around: A bioinspired approach to visual attention through object motion sensitivityCode0
Layered Motion Fusion: Lifting Motion Segmentation to 3D in Egocentric Videos—0
Deep Learning and Hybrid Approaches for Dynamic Scene Analysis, Object Detection and Motion Tracking—0
On Moving Object Segmentation from Monocular Video with Transformers—0
RoMo: Robust Motion Segmentation Improves Structure from Motion—0
DATAP-SfM: Dynamic-Aware Tracking Any Point for Robust Structure from Motion in the Wild—0
Human Insights Driven Latent Space for Different Driving Perspectives: A Unified Encoder for Efficient Multi-Task Inference—0
CV-MOS: A Cross-View Model for Motion SegmentationCode0
Retina-Inspired Object Motion Segmentation for Event-Cameras—0
Hardware-Algorithm Re-engineering of Retinal Circuit for Intelligent Object Motion Segmentation—0
Show:102550
← PrevPage 1 of 5Next →
All datasetsApolloScapeHopkins155KT3DMoSegMTPV62

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Rule BasedAccuracy90—Unverified
2Rel-Att-GCNAccuracy89—Unverified
3MRGCNAccuracy86—Unverified
4MRGCN-LSTMAccuracy72—Unverified
5St-RNNAccuracy63—Unverified
#ModelMetricClaimedVerifiedStatus
1SSCClassification Error2.18—Unverified
2T-LinkageClassification Error1.97—Unverified
3RSIMClassification Error1.01—Unverified
4MVCClassification Error0.31—Unverified
#ModelMetricClaimedVerifiedStatus
1MultiViewClusteringError7.92—Unverified
#ModelMetricClaimedVerifiedStatus
1MVCClassification Error0.65—Unverified