GraphDeco

Cotemporal Multi-View Video Segmentation
Presentation | Team members | Collaborations | Publications | Job offers | Contact

 

Cotemporal Multi-View Video Segmentation

International Conference on 3D Vision (3DV) - 2016
Download the publication : final_umvseg_3dv.pdf [1.9Mo]  
We address the problem of multi-view video segmenta-tion of dynamic scenes in general and outdoor environments with possibly moving cameras. Multi-view methods for dynamic scenes usually rely on geometric calibration to impose spatial shape constraints between viewpoints. In this paper, we show that the calibration constraint can be relaxed while still getting competitive segmentation results using multi-view constraints. We introduce new multi-view cotemporality constraints through motion correlation cues, in addition to common appearance features used by co-segmentation methods to identify co-instances of objects. We also take advantage of learning based segmentation strategies by casting the problem as the selection of monoc-ular proposals that satisfy multi-view constraints. This yields a fully automated method that can segment subjects of interest without any particular pre-processing stage. Results on several challenging outdoor datasets demonstrate the feasibility and robustness of our approach.

Images and movies

 

See also

Acknowledgements and Funding

The authors are grateful for the support of EU project CR-PLAY (no 611089) www.cr-play.eu and project SEMAPOLIS (ANR-13-CORD-0003).

BibTex references

@InProceedings{DFBPD16,
  author       = "Djelouah, Abdelaziz and Franco, Jean-S\'ebastien and Boyer, Edmond and P\'erez, Patrick and Drettakis, George",
  title        = "Cotemporal Multi-View Video Segmentation",
  booktitle    = "International Conference on 3D Vision (3DV)",
  year         = "2016",
  url          = "http://www-sop.inria.fr/reves/Basilic/2016/DFBPD16"
}

Other publications in the database

» Abdelaziz Djelouah
» George Drettakis