Skip to Main content Skip to Navigation
Conference papers

Superpixel-based depth map inpainting for RGB-D view synthesis

Pierre Buyssens 1 Maxime Daisy 1 David Tschumperlé 1 Olivier Lézoray 1
1 Equipe Image - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : In this paper we propose an approach to inpaint holes in depth maps that appear when synthesizing virtual views from a RGB-D scenes. Based on a superpixel oversegmentation of both the original and synthesized views, the proposed approach efficiently deals with many occlusion situations where most of previous approaches fail. The use of superpixels makes the algorithm more robust to inaccurate depth maps, while giving an efficient way to model the image. Extensive comparisons to relevant state-of-the-art methods show that our approach outperforms qualitatively and quantitavely these existing approaches.
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download
Contributor : Pierre Buyssens <>
Submitted on : Wednesday, May 20, 2015 - 2:36:45 PM
Last modification on : Wednesday, November 13, 2019 - 3:40:33 PM
Long-term archiving on: : Tuesday, September 15, 2015 - 6:28:42 AM


Files produced by the author(s)


  • HAL Id : hal-01153769, version 1


Pierre Buyssens, Maxime Daisy, David Tschumperlé, Olivier Lézoray. Superpixel-based depth map inpainting for RGB-D view synthesis. International Conference on Image Processing, Sep 2015, Québec City, Canada. ⟨hal-01153769⟩



Record views


Files downloads