Kinecting the dots: Particle based scene flow from depth sensors
Citations Over TimeTop 10% of 2011 papers
Abstract
The motion field of a scene can be used for object segmentation and to provide features for classification tasks like action recognition. Scene flow is the full 3D motion field of the scene, and is more difficult to estimate than it's 2D counterpart, optical flow. Current approaches use a smoothness cost for regularisation, which tends to over-smooth at object boundaries. This paper presents a novel formulation for scene flow estimation, a collection of moving points in 3D space, modelled using a particle filter that supports multiple hypotheses and does not oversmooth the motion field. In addition, this paper is the first to address scene flow estimation, while making use of modern depth sensors and monocular appearance images, rather than traditional multi-viewpoint rigs. The algorithm is applied to an existing scene flow dataset, where it achieves comparable results to approaches utilising multiple views, while taking a fraction of the time.
Related Papers
- → The computation of optical flow(1995)1,251 cited
- → Optical Flow Estimation versus Motion Estimation(2012)4 cited
- → <title>Point-constrained optical flow for LV motion detection</title>(2000)12 cited
- A New Method of 3D Motion and Structure Estimation Based on Spare Optical Flow(2003)
- → Describing Motion of Dynamic Objects for a Moving Camera(2022)1 cited