Skip to Main content Skip to Navigation
Conference papers

Incomplete 3D Motion Trajectory Segmentation and 2D-to-3D Label Transfer for Dynamic Scene Analysis

Abstract : The knowledge of the static scene parts and the moving objects in a dynamic scene plays a vital role for scene modelling, understanding, and landmark-based robot navigation. The key information for these tasks lies on semantic labels of the scene parts and the motion trajectories of the dynamic objects. In this work, we propose a method that segments the 3D feature trajectories based on their motion behaviours, and assigns them semantic labels using 2D-to-3D label transfer. These feature trajectories are constructed by using the proposed trajectory recovery algorithm which takes the loss of feature tracking into account. We introduce a complete framework for static-map and dynamic objects' reconstruction, as well as semantic scene understanding for a calibrated and moving 2D-3D camera setup. Our motion segmentation approach is faster by two orders of magnitude, while performing better than the state-of-the-art 3D motion segmentation methods, and successfully handles the previously discarded incomplete trajectory scenarios.
Complete list of metadatas

Cited literature [32 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01569325
Contributor : Cansen Jiang <>
Submitted on : Wednesday, July 26, 2017 - 3:08:38 PM
Last modification on : Monday, March 30, 2020 - 8:42:19 AM

File

IROS17_Camera_Ready.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01569325, version 1

Citation

Cansen Jiang, Danda Paudel, Yohan Fougerolle, David Fofi, Cédric Demonceaux. Incomplete 3D Motion Trajectory Segmentation and 2D-to-3D Label Transfer for Dynamic Scene Analysis. IEEE/RSJ International Conference on Intelligent Robots and Systems - IROS, Sep 2017, Vancouver, Canada. ⟨hal-01569325⟩

Share

Metrics

Record views

376

Files downloads

655