Variational Image-Based Rendering with Gradient Constraints

Grégoire Nieto 1 Frédéric Devernay 1 James L. Crowley 2
1 IMAGINE - Intuitive Modeling and Animation for Interactive Graphics & Narrative Environments
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
2 PERVASIVE INTERACTION
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
Abstract : Multi-view image-based rendering consists in generating a novel view of a scene from a set of source views. In general, this works by first doing a coarse 3D reconstruction of the scene, and then using this reconstruction to establish correspondences between source and target views, followed by blending the warped views to get the final image. Unfortunately, discontinuities in the blending weights, due to scene geometry or camera placement, result in artifacts in the target view. In this paper, we show how to avoid these artifacts by imposing additional constraints on the image gradients of the novel view. We propose a variational framework in which an energy functional is derived and optimized by iteratively solving a linear system. We demonstrate this method on several structured and unstructured multi-view datasets, and show that it numerically outperforms state-of-the-art methods, and eliminates artifacts that result from visibility discontinuities.
Complete list of metadatas

Cited literature [27 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01402528
Contributor : Nieto Grégoire <>
Submitted on : Thursday, November 24, 2016 - 5:46:47 PM
Last modification on : Friday, April 26, 2019 - 6:38:02 PM
Long-term archiving on : Monday, March 20, 2017 - 7:32:39 PM

File

ic3d2016.pdf
Files produced by the author(s)

Identifiers

Citation

Grégoire Nieto, Frédéric Devernay, James L. Crowley. Variational Image-Based Rendering with Gradient Constraints. IC3D - 2016 International Conference on 3D Imaging, Dec 2016, Liège, Belgium. pp.1-8, ⟨10.1109/IC3D.2016.7823449⟩. ⟨hal-01402528⟩

Share

Metrics

Record views

734

Files downloads

613