Scene Flow Estimation
78 papers with code • 5 benchmarks • 8 datasets
Optical flow is a two-dimensional motion field in the image plane. It is the projection of the three-dimensional motion of the world. If the world is completely non-rigid, the motions of the points in the scene may all be indepen- dent of each other. One representation of the scene motion is therefore a dense three-dimensional vector field defined for every point on every surface in the scene. By analogy with optical flow, we refer to this three-dimensional motion field as scene flow.
Source: Vedula, Sundar, et al. "Three-dimensional scene flow." IEEE transactions on pattern analysis and machine intelligence 27.3 (2005): 475-480. pdf
Benchmarks
These leaderboards are used to track progress in Scene Flow Estimation
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
KITTI
FlyingThings3D
Argoverse 2
DeformingThings4D
Spring
VBR
Middlebury 2003
EKubric
Most implemented papers
Scalable Scene Flow from Point Clouds in the Real World
kth-rpl/deflow
•
•
In this work, we introduce a new large-scale dataset for scene flow estimation derived from corresponding tracked 3D objects, which is $\sim$1, 000$\times$ larger than previous real-world datasets in terms of the number of annotated frames.
A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation
HKBU-HPML/FADNet
•
•
CVPR 2016
By combining a flow and disparity estimation network and training it jointly, we demonstrate the first scene flow estimation with a convolutional network.
DeFlow: Decoder of Scene Flow Network in Autonomous Driving
kth-rpl/deflow
•
•
Scene flow estimation determines a scene's 3D motion field, by predicting the motion of points in the scene, especially for aiding tasks in autonomous driving.
HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds
laoreja/HPLFlowNet
•
•
CVPR 2019
We present a novel deep neural network architecture for end-to-end scene flow estimation that directly operates on large-scale 3D point clouds.
MeteorNet: Deep Learning on Dynamic 3D Point Cloud Sequences
xingyul/meteornet
•
•
ICCV 2019
Understanding dynamic 3D environment is crucial for robotic agents and many other applications.
PointPWC-Net: A Coarse-to-Fine Network for Supervised and Self-Supervised Scene Flow Estimation on 3D Point Clouds
We propose a novel end-to-end deep scene flow model, called PointPWC-Net, on 3D point clouds in a coarse-to-fine fashion.
Accurate Point Cloud Registration with Robust Optimal Transport
uncbiag/shapmagn
•
•
NeurIPS 2021
Finally, we showcase the performance of transport-enhanced registration models on a wide range of challenging tasks: rigid registration for partial shapes; scene flow estimation on the Kitti dataset; and nonparametric registration of lung vascular trees between inspiration and expiration.
Self-Supervised Scene Flow Estimation with 4-D Automotive Radar
toytiny/raflow
•
•
Scene flow allows autonomous vehicles to reason about the arbitrary motion of multiple independent objects which is the key to long-term mobile autonomy.
3D Object Detection with a Self-supervised Lidar Scene Flow Backbone
emecercelik/ssl-3d-detection
•
•
Our main contribution leverages learned flow and motion representations and combines a self-supervised backbone with a supervised 3D detection head.
Spring: A High-Resolution High-Detail Dataset and Benchmark for Scene Flow, Optical Flow and Stereo
While recent methods for motion and stereo estimation recover an unprecedented amount of details, such highly detailed structures are neither adequately reflected in the data of existing benchmarks nor their evaluation methodology.
