Mumtaz Sheikh

Bio: Mumtaz Sheikh is an academic researcher from Lahore University of Management Sciences. The author has contributed to research in topics: Silicon carbide & Lens (optics). The author has an hindex of 9, co-authored 28 publications receiving 493 citations. Previous affiliations of Mumtaz Sheikh include University of Central Florida.

Exploring the space of a human action

Abstract: One of the fundamental challenges of recognizing actions is accounting for the variability that arises when arbitrary cameras capture humans performing actions. In this paper, we explicitly identify three important sources of variability: (1) viewpoint, (2) execution rate, and (3) anthropometry of actors, and propose a model of human actions that allows us to investigate all three. Our hypothesis is that the variability associated with the execution of an action can be closely approximated by a linear combination of action bases in joint spatio-temporal space. We demonstrate that such a model bounds the rank of a matrix of image measurements and that this bound can be used to achieve recognition of actions based only on imaged data. A test employing principal angles between subspaces that is robust to statistical fluctuations in measurement data is presented to find the membership of an instance of an action. The algorithm is applied to recognize several actions, and promising results have been obtained.

Hybrid Wireless-Wired Optical Sensor for Extreme Temperature Measurement in Next Generation Energy Efficient Gas Turbines

Abstract: Accuracy, reliability, and long lifetimes are critical parameters for sensors measuring temperature in gas turbines of clean coal-fired power plants. Greener high efficiency next generation power plants need gas turbines operating at extremely high temperatures of 1500°C, where present thermocouple temperature probe technology fails to operate with reliable and accurate readings over long lifetimes. To solve this pressing problem, we have proposed the concept of a new hybrid class of all-silicon carbide (SiC) optical sensor, where a single crystal SiC optical chip is embedded in a sintered SiC tube assembly, forming a coefficient of thermal expansion (CTE) matched all-SiC front-end probe. Because chip and host material are CTE matched, optimal handling of extreme thermal ramps and temperatures is possible. In this article, we demonstrate the first successful industrial combustor rig test of this hybrid all-SiC temperature sensor front-end probe indicating demonstrated probe structural robustness to 1600°C and rig test data to ~1200°C. The design of the rig test sensor system is presented and data are analyzed.

All-silicon carbide hybrid wireless-wired optics temperature sensor network basic design engineering for power plant gas turbines

Abstract: Proposed is a novel design of a fiber-remoted temperature sensor network for operation in the extreme environments of power generation gas turbines. The network utilizes a robust all-Silicon Carbide wireless-wired hybrid temperature probe design that features an all-passive front-end, active laser beam targeting, and the use of an optical wedge that eliminates optical interferometric noise in addition to serving as a partial vacuum window for the probe cavity to minimize laser beam wander due to air turbulence. An example basic network is built at the 1550 nm band using 1 × 2 micro-electro-mechanical systems (MEMS) fiber-optic switches with engineered sensor system robust performance observed at 1000°C using a custom assembled all-SiC probe with a Magnesium Fluoride (MgF2) high temperature window.

Motion-free hybrid design laser beam propagation analyzer using a digital micromirror device and a variable focus liquid lens

Abstract: To the best of our knowledge, we propose the first motion-free laser beam propagation analyzer with a hybrid design using a digital micromirror device (DMD) and a liquid electronically controlled variable focus lens (ECVFL). Unlike prior analyzers that require profiling the beam at multiple locations along the light propagation axis, the proposed analyzer profiles the beam at the same plane for multiple values of the ECVFL focal length, thus eliminating beam profiler assembly motion. In addition to measuring standard Gaussian beam parameters, the analyzer can also be used to measure the M2 beam propagation parameter of a multimode beam. Proof-of-concept beam parameter measurements with the proposed analyzer are successfully conducted for a 633 nm laser beam. Given the all-digital nature of the DMD-based profiling and all-analog motion-free nature of the ECVFL beam focus control, the proposed analyzer versus prior art promises better repeatability, speed, and reliability.

Demonstration of Pinhole Laser Beam Profiling Using a Digital Micromirror Device

Abstract: Demonstrated for the first time, to the best of the authors' knowledge, is two-dimensional (2-D) pinhole laser beam profiling using a Texas Instruments visible band digital micromirror device (DMD). Software controlled micromirror digital tilt positions across the DMD plane create a moving pinhole for sampling an arbitrary distribution laser beam power profile. A 532-nm 0.5-W laser coupled to an optically addressed nematic liquid crystal spatial light modulator is used to generate a laser beam black-and-white high-resolution test line pattern having a 62.5-mum linewidth. The test pattern is successfully profiled using a DMD formed 27.36 mum times 27.36 mum pinhole. Demonstrated for the first time is 2-D knife-edge DMD-based profiling of an ultraviolet 337-nm 4-ns pulsewidth 10-Hz pulsed laser beam.

Human activity analysis: A review

Abstract: Human activity recognition is an important area of computer vision research. Its applications include surveillance systems, patient monitoring systems, and a variety of systems that involve interactions between persons and electronic devices such as human-computer interfaces. Most of these applications require an automated recognition of high-level activities, composed of multiple simple (or atomic) actions of persons. This article provides a detailed overview of various state-of-the-art research papers on human activity recognition. We discuss both the methodologies developed for simple human actions and those for high-level activities. An approach-based taxonomy is chosen that compares the advantages and limitations of each approach. Recognition methodologies for an analysis of the simple actions of a single person are first presented in the article. Space-time volume approaches and sequential approaches that represent and recognize activities directly from input images are discussed. Next, hierarchical recognition methodologies for high-level activities are presented and compared. Statistical approaches, syntactic approaches, and description-based approaches for hierarchical recognition are discussed in the article. In addition, we further discuss the papers on the recognition of human-object interactions and group activities. Public datasets designed for the evaluation of the recognition methodologies are illustrated in our article as well, comparing the methodologies' performances. This review will provide the impetus for future research in more productive areas.

Human Action Recognition by Representing 3D Skeletons as Points in a Lie Group

Abstract: Recently introduced cost-effective depth sensors coupled with the real-time skeleton estimation algorithm of Shotton et al. [16] have generated a renewed interest in skeleton-based human action recognition. Most of the existing skeleton-based approaches use either the joint locations or the joint angles to represent a human skeleton. In this paper, we propose a new skeletal representation that explicitly models the 3D geometric relationships between various body parts using rotations and translations in 3D space. Since 3D rigid body motions are members of the special Euclidean group SE(3), the proposed skeletal representation lies in the Lie group SE(3)×…×SE(3), which is a curved manifold. Using the proposed representation, human actions can be modeled as curves in this Lie group. Since classification of curves in this Lie group is not an easy task, we map the action curves from the Lie group to its Lie algebra, which is a vector space. We then perform classification using a combination of dynamic time warping, Fourier temporal pyramid representation and linear SVM. Experimental results on three action datasets show that the proposed representation performs better than many existing skeletal representations. The proposed approach also outperforms various state-of-the-art skeleton-based human action recognition approaches.

Machine Recognition of Human Activities: A Survey

Abstract: The past decade has witnessed a rapid proliferation of video cameras in all walks of life and has resulted in a tremendous explosion of video content. Several applications such as content-based video annotation and retrieval, highlight extraction and video summarization require recognition of the activities occurring in the video. The analysis of human activities in videos is an area with increasingly important consequences from security and surveillance to entertainment and personal archiving. Several challenges at various levels of processing-robustness against errors in low-level processing, view and rate-invariant representations at midlevel processing and semantic representation of human activities at higher level processing-make this problem hard to solve. In this review paper, we present a comprehensive survey of efforts in the past couple of decades to address the problems of representation, recognition, and learning of human activities from video and related applications. We discuss the problem at two major levels of complexity: 1) "actions" and 2) "activities." "Actions" are characterized by simple motion patterns typically executed by a single human. "Activities" are more complex and involve coordinated actions among a small number of humans. We will discuss several approaches and classify them according to their ability to handle varying degrees of complexity as interpreted above. We begin with a discussion of approaches to model the simplest of action classes known as atomic or primitive actions that do not require sophisticated dynamical modeling. Then, methods to model actions with more complex dynamics are discussed. The discussion then leads naturally to methods for higher level representation of complex activities.