Image Understanding at the GRASP Laboratory
نویسنده
چکیده
Research in the GRASP Laboratory has two main themes, parameterized multi-dimensional segmentation and robust decision making under uncertainty. The multi-dimensional approach interweaves segmentation with representation. The data is explained as a best fit in view of parametric primitives. These primitives are based on physical and geometric properties of objects and are limited in number. We use primitives at the volumetric level, the surface level, and the occluding contour level, and combine the results. The robust decision making allows us to combine data from multiple sensors. Sensor measurements have bounds based on the physical limitations of the sensors. We use this information without making a priori assumptions of distributions within the intervals or a priori assumptions of the probability of a given result.
منابع مشابه
The MASKS Face Image Data Set: Collection Procedures and Progress Report
In this report, we describe the creation of a new face image data set, an effort undertaken as part of the MASKS (Maintaining Anonymity by Sequestering Key Statistics) Project. This set of high-quality, carefully controlled images were taken from April until December 2006 at the GRASP (General Robotics, Automation, Sensing and Perception) Laboratory. Comments University of Pennsylvania Departme...
متن کاملObject-Oriented Method for Automatic Extraction of Road from High Resolution Satellite Images
As the information carried in a high spatial resolution image is not represented by single pixels but by meaningful image objects, which include the association of multiple pixels and their mutual relations, the object based method has become one of the most commonly used strategies for the processing of high resolution imagery. This processing comprises two fundamental and critical steps towar...
متن کاملOn the deformation of image intensity and zero-crossing contours under motion
Image intensity and edge are two major sources of information for estimating the motion in the image plane. The 2-D motion obtained by analyzing the deformation of intensity and/or edges is used to recover the 3-D motion and structure. In this paper we show that the motion defined by the image intensity differs from the motion revealed by the (zerocrossing) edge. Understanding of this discrepan...
متن کاملDiscrimintive Image Warping with Attribute Flow
We address the problem of finding deformation between two images for the purpose of recognizing objects. The challenge is that discriminative features are often transformation-variant (e.g. histogram of oriented gradients, texture), while transformation-invariant features (e.g. intensity, color) are often not discriminative. We introduce the concept of attribute flow which explicitly models how...
متن کاملGRASP NEWS: Volume 7, Number 1
Since its beginning in 1983, the GRASP News has chronicled the research efforts of the Grasp Laboratory. This edition, which covers developments for the year 1990, follows the format of previous editions. The Feature Article, however, in a departure from tradition does not highlight a particular research project. Helen Anderson presents a summary of the Laboratory research projects and activiti...
متن کامل