AbstractWe propose a novel algorithm for monocular depth estimation that decomposes metric map into normalized and scale features. The proposed network is composed of shared encoder three decoders, called G-Net, N-Net, M-Net, which estimate gradient maps, map, respectively. M-Net learns to depths more accurately using relative features extracted by G-Net N-Net. has the advantage it can use data...

Multiview video in “texture-plus-depth” format enables decoder to synthesize freely chosen intermediate views for enhanced visual experience. Nevertheless, transmission of multiple texture and depth maps over bandwidthconstrained and loss-prone networks is challenging, especially for conferencing applications with stringent deadlines. In this paper, we examine the problem of loss-resilient codi...

We present a system for automatically generating custom, structured image-maps for input depth-maps. Our system thus allows quick fitting of masses of objects with tailor-made image-maps. Given a depth-map of a novel 3D object, our method tiles it with intensities from similar, pre-collected, textured objects. These are seamlessly merged to form the new image-map. This process is performed by o...

In this paper, we propose a method to extract depth from motion, texture and intensity. We first analyze the depth map to extract a set of depth cues. Then, based on these depth cues, we process the colored reference video, using texture, motion, luminance and chrominance content, to extract the depth map. The processing of each channel in the Y CRCB−color space is conducted separately. We test...

The profile of a 10 mm wide and 3μm deep grinding imprint is as unique as a human fingerprint. To utilize this for fingerprinting mechanical components, a robust and strong characterization has to be used. We propose a feature-based approach, in which features of a 1D profile are detected and described in its 2D space-frequency representation. We show that the approach is robust on depth maps a...

A single two dimensional (2D) image does not contain depth information. An infinite number of points in the three dimensional (3D) space are projected to the same point in the image plane. But a single 2D image has some monocular depth cues, by which we can make a hypothesis of depth variation in the image to generate a depth map. This paper proposes an interactive method of depth map generatio...

We present a novel enhancement method that addresses the problem of corrupted edge information in depth maps. Corrupted depth information manifests itself in zigzag edges instead of straight ones. We extract the depth information from an associated color stream and use this information to enhance the original depth map. Besides the visual results, a quantitative analysis is conducted to prove t...