A scale space for texture+depth images based on a discrete Laplacian operator

Maxim Karpushin (1) , Giuseppe Valenzise (1) , Frédéric Dufaux (1)

1 LTCI - Laboratoire Traitement et Communication de l'Information

In this paper we design a smoothing filter for texture+depth images based on anisotropic diffusion. Our proposed filter enables to generate a scale space on the texture image guided by depth information , and is linear and numerically stable. We show experimentally that using scene geometry preserves the internal structure of 3D surfaces (e.g., it avoids smoothing across object boundaries). As a consequence, the result of smoothing is more independent to changes in the camera position. To illustrate the practical utility of a scale space with such properties, we integrate our filter into the SIFT key-point detector, getting a substantial improvement of the repeatability of detected keypoints under significant viewpoint position changes.

Mots clés: image smoothing; texture+depth; anisotropic diffusion; handcrafted features; keypoint repeatability

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement des images [eess.IV]

Maxim Karpushin, Giuseppe Valenzise, Frédéric Dufaux. A scale space for texture+depth images based on a discrete Laplacian operator. IEEE International Conference on Multimedia and Expo, Jun 2015, Turin, Italy. ⟨10.1109/ICME.2015.7177500⟩. ⟨hal-01299841⟩