In this paper, detection of pedestrian groups and counting of the number of pedestrians in each group using “subtraction stereo” are discussed. Subtraction stereo is a stereo vision method that focuses on the movement of objects to make a stereo camera robust and produces range images for moving regions. Pedestrian groups are detected with a standard labeling, and three dimensional (3D) feature...

We present a novel stereo image denoising algorithm. Our algorithm takes as an input a pair of noisy images of an object captured form two different directions. We use the structural similarity index as a similarity metric for identifying locations of similar patches in the input images. We adapt the Non-Local Means algorithm for denoising collected patches from the input images. We validate ou...

We present a stereo image denoising algorithm. Our algorithm takes as an input a pair of noisy images of an object captured from two different directions (stereo images). We use either Maximum Difference or Singular Value Decomposition similarity metrics for identifying locations of similar searching windows in the input images. We adapt the Non-local Means algorithm for denoising collected pat...

This report describes a dataset of color, stereo images of ordnance collected at a live-re test range near Nellis Air Force Base. The images consist primarily of BLU-97 shells. 174 stereo pairs of images were collected, containing approximately 184 instances of BLU-97. The data was collected using a pair of Kodak DCS-410 digital cameras mounted rigidly on a tripod. Subsequent processing was per...

In this paper, a new approach for optimizing disparity candidates space is proposed for the solution of dense stereo matching problem. The main objectives of this approachare the reduction of average number of disparity candidates per pixel with low computational cost and high assurance of retaining the correct answer. These can be realized due to the effective use of multiple radial windows, i...

This PhD aims at the design and implementation of real-time hardware-based stereo vision systems. Stereo vision deals with images acquired by a stereo camera setup, where the disparity between the stereo images allows depth estimation within a scene. In the first part of my research a new hardwareefficient real time disparity map computation module was developed which involved a fully parallel-...

Real-time active 3D range cameras based on time-of-flight technology using the Photonic Mixer Device (PMD) can be considered as a complementary technique for stereo-vision based depth estimation. Since those systems directly yield 3D measurements, they can also be used for initializing vision based approaches, especially in highly dynamic environments. Fusion of PMD depth images with passive in...

Orbital stereo data is acquired by covering same area from two exposure stations in space with oblique viewing capability. It is used for generating digital elevation models (DEM) and three dimensional viewing of the terrain in a softcopy photogrammetric workstation. A good base to height ratio (B/H), usually greater than 0.5, is necessary for generation of good quality absolute DEM. Processing...

In order to perform localization and navigation over significant distances (up to a few kilometers), it is important to be able to accurately map the terrain to be traversed. Local methods, such as conventional stereo, do not scale well to large distances and prevent longrange planning. In this paper, we discuss wide-baseline stereo techniques for rovers. In wide-baseline stereo, the image pair...

Estimating the pose and 3D shape of a large variety of instances within an object class from stereo images is a challenging problem, especially in realistic conditions such as urban street scenes. We propose a novel approach for using compact shape manifolds of the shape within an object class for object segmentation, pose and shape estimation. Our method first detects objects and estimates the...