The detection of object boundaries is a critical first step for many visual processing tasks.
Multiple cues (we consider luminance, color, motion and binocular disparity) available in the …

There is an emerging sense that the vulnerability of Image Convolutional Neural Networks
(CNN), ie, sensitivity to image corruptions, perturbations, and adversarial attacks, is …

We propose a novel framework for contour based object detection and recognition, which
we formulate as a joint contour fragment grou** and labeling problem. For a given set of …

Interest point-based multiview 3D reconstruction and calibration methods have been very
successful in select applications but are not applicable when an abundance of feature points …

The raising standards in manufacturing demands reliable and fast industrial quality control
mechanisms. This paper proposes an accurate, yet easy to install multi-view, close range …

Reconstructing 3D scenes from multiple views has made impressive strides in recent years,
chiefly by correlating isolated feature points, intensity patterns, or curvilinear structures. In …

Content generated by users is one of the most interesting phenomena of published media.
However, the possibility of unrestricted edition is a source of doubts about its quality. This …

The field of multiple view geometry has seen tremendous progress in reconstruction and
calibration due to methods for extracting reliable point features and key developments in …

In this paper we propose a novel framework for contour based object detection from
cluttered environments. Given a contour model for a class of objects, it is first decomposed …

Proposes a novel scheduling framework for a real-time environment that experiences
dynamic changes. This framework is capable of adjusting the system workload in …