Depth map estimation is a prerequisite for useful plenoptic rendering algorithms, such as all-in-focus rendering and depth-based matting. We provide a simple GPU-based algorithm that estimates the depth of each pixel in the rendered image as well as the uncertainty of that estimate. We perform several rendering experiments using depth information. Finally, explore what factors harm the accuracy...

As demand for advanced photographic applications on hand-held devices grows, these electronics require the capture of high quality depth. However, under low-light conditions, most devices still suffer from low imaging quality and inaccurate depth acquisition. To address the problem, we present a robust depth estimation method from a short burst shot with varied intensity (i.e., Auto Bracketing)...

Stereoscopic images are hard to get right, and comfortable images are often only produced after repeated trial and error. The main difficulty is controlling the stereoscopic camera parameters so that the viewer does not experience eye strain or double images from excessive perceived depth. Additionally, for head tracked displays, the perceived objects can distort as the viewer moves to look aro...

This paper focuses on real-time markerless motion capture. The body pose of a person is estimated from depth images using an Iterative Closest Point algorithm. We present a very efficient approach, that estimates up to 28 degrees of freedom from 1000 data points with 4Hz. This is achieved by nonlinear optimization techniques using an analytically derived Jacobian and highly optimized correspond...

To recover depth from images, the human visual system uses many monocular depth cues, which vision research has only begun to explore. Because a given image can have many possible interpretations, constraints are needed to eliminate ambiguity, and the most powerful constraints are domain specific. As an experiment in the automatic discovery and exploitation of constraints, Genetic Programming w...

This paper proposes an algorithm for inferring a 3D mesh using the robust cost function proposed by Ruttle et al. [12]. Our contribution is in proposing a new algorithm for inference that is very suitable for parallel architecture. The cost function also provides a goodness of fit for each element of the mesh which is correlated to the distance to the ground truth, hence providing informative f...

5.1 INTRODUCTION This chapter addresses the problem of depth estimation using a sequence of images. In the previous chapter, we have eliminated the correspondence problem by taking two specific orthogonal views. In this chapter, this limitation is sought to be removed by employing point correspondence. The equation of motion derived from spherical projection will be reformulated so as to apply ...

This paper proposes to infer accurately a 3D shape of an object captured by a depth camera from multiple view points. The Generalised Relaxed Radon Transform (GR2T) [1] is used here to merge all depth images in a robust kernel density estimate that models the surface of an object in the 3D space. The kernel is tailored to capture the uncertainty associated with each pixel in the depth images. T...

We focus on the challenging problem of efficient mouse 3D pose estimation based on static images, and especially single depth images. We introduce an approach to discriminatively train the split nodes of trees in random forest to improve their performance on estimation of 3D joint positions of mouse. Our algorithm is capable of working with different types of rodents and with different types of...

We present a strategy that combines color and depth images to detect people in indoor environments. Similarity of image appearance and closeness in 3D position over time yield weights on the edges of a directed graph that we partition greedily into tracklets, sequences of chronologically ordered observations with high edge weights. Each tracklet is assigned the highest score that a Histograms-o...