Scott McCloskey

Bio: Scott McCloskey is an academic researcher from Honeywell. The author has contributed to research in topics: Shutter & Motion blur. The author has an hindex of 27, co-authored 107 publications receiving 3444 citations. Previous affiliations of Scott McCloskey include McGill University.

Integrated dimensioning and weighing system

Abstract: An object analysis system includes a scale for measuring the weight of the object, a range camera configured to produce a range image of an area in which the object is located, and a computing device configured to determine the dimensions of the object based, at least in part, on the range image. Methods for determining the dimensions of an object include capturing a range image and/or a visible image of a scene that includes the object.

System and method for package dimensioning

Abstract: A system and method for package dimensioning is provided. The package-dimensioning system includes an image capturing subsystem for acquiring information about an object within the image-capturing subsystem's field of view. A features-computation module analyzes object information and compiles a feature set describing the object's surface features. A classification module analyzes the feature set and categorizes the object's shape. A shape-estimation module estimates the dimensions of the object.

Dimensioning system with guided alignment

Abstract: A dimensioning system including a computing device running an alignment software program is disclosed. The alignment software uses range information from a range sensor in order to generate alignment messages. The alignment messages may help a user define a frame of reference and align the dimensioning system's range sensor for improved dimensioning performance.

Variable depth of field barcode scanner

Abstract: Generally discussed herein are techniques, software, apparatuses, and systems configured for variable depth of field scanners. In one or more embodiments, a method can include exposing a light sensor of a barcode scanner for a specified time, altering a focal point of a variable focus element situated in the light path of the light sensor from a first focal point to a second focal point in the specified time, and processing an image of a barcode produced by the light sensor to deblur the image.

Indicia reading terminals and methods for decoding decodable indicia employing light field imaging

Abstract: A terminal for decoding decodable indicia includes a plenoptic imaging subsystem comprising an image sensor array and plenoptic imaging optics operable to project a plenoptic image of a space containing the decodable indicia onto the image sensor array, a hand held housing encapsulating a least a portion of the plenoptic imaging subsystem, a trigger for initiating operation of the plenoptic imaging subsystem to obtain plenoptic image data of the decodable indicia, and an illumination source for projecting illumination onto the decodable indicia; an aimer for projecting an aimer pattern onto the decodable indicia. The terminal is operable, responsive to detecting that the trigger has been actuated by an operator, to obtain plenoptic image data from the image sensor array, to obtain first rendered image data based on at least a portion of the plenoptic image data, and to attempt to decode the decodable indicia represented in the rendered image data.

Analyzing and Improving the Image Quality of StyleGAN

Abstract: The style-based GAN architecture (StyleGAN) yields state-of-the-art results in data-driven unconditional generative image modeling. We expose and analyze several of its characteristic artifacts, and propose changes in both model architecture and training methods to address them. In particular, we redesign the generator normalization, revisit progressive growing, and regularize the generator to encourage good conditioning in the mapping from latent codes to images. In addition to improving image quality, this path length regularizer yields the additional benefit that the generator becomes significantly easier to invert. This makes it possible to reliably attribute a generated image to a particular network. We furthermore visualize how well the generator utilizes its output resolution, and identify a capacity problem, motivating us to train larger models for additional quality improvements. Overall, our improved model redefines the state of the art in unconditional image modeling, both in terms of existing distribution quality metrics as well as perceived image quality.

Analyzing and Improving the Image Quality of StyleGAN

Abstract: The style-based GAN architecture (StyleGAN) yields state-of-the-art results in data-driven unconditional generative image modeling. We expose and analyze several of its characteristic artifacts, and propose changes in both model architecture and training methods to address them. In particular, we redesign the generator normalization, revisit progressive growing, and regularize the generator to encourage good conditioning in the mapping from latent codes to images. In addition to improving image quality, this path length regularizer yields the additional benefit that the generator becomes significantly easier to invert. This makes it possible to reliably attribute a generated image to a particular network. We furthermore visualize how well the generator utilizes its output resolution, and identify a capacity problem, motivating us to train larger models for additional quality improvements. Overall, our improved model redefines the state of the art in unconditional image modeling, both in terms of existing distribution quality metrics as well as perceived image quality.

A Survey on Multi-view Learning

Abstract: In recent years, a great many methods of learning from multi-view data by considering the diversity of different views have been proposed. These views may be obtained from multiple sources or different feature subsets. In trying to organize and highlight similarities and differences between the variety of multi-view learning approaches, we review a number of representative multi-view learning algorithms in different areas and classify them into three groups: 1) co-training, 2) multiple kernel learning, and 3) subspace learning. Notably, co-training style algorithms train alternately to maximize the mutual agreement on two distinct views of the data; multiple kernel learning algorithms exploit kernels that naturally correspond to different views and combine kernels either linearly or non-linearly to improve learning performance; and subspace learning algorithms aim to obtain a latent subspace shared by multiple views by assuming that the input views are generated from this latent subspace. Though there is significant variance in the approaches to integrating multiple views to improve learning performance, they mainly exploit either the consensus principle or the complementary principle to ensure the success of multi-view learning. Since accessing multiple views is the fundament of multi-view learning, with the exception of study on learning a model from multiple views, it is also valuable to study how to construct multiple views and how to evaluate these views. Overall, by exploring the consistency and complementary properties of different views, multi-view learning is rendered more effective, more promising, and has better generalization ability than single-view learning.