Outdoor scene parsing models are often trained on ideal datasets and produce quality results. However, this leads to a discrepancy when applied to the real world. The quality of scene parsing, particularly sky classification, decreases in night time images, images involving varying weather conditions, and scene changes due to seasonal weather. This project focuses on approaching these challenges by using a state-of-the-art model in conjunction with non-ideal datasets: SkyFinder and a subset from the SUN database containing the Sky object. We focus specifically on sky segmentation, the task of determining sky and not-sky pixels, and improving upon an existing state-of-the-art model: Re-fineNet. As a result of our efforts, we have seen an improvement of 10-15% in the average MCR compared to the prior methods on the SkyFinder dataset. We have also improved from an off-the-shelf model in terms of average mIOU by nearly 35%. Further, we analyze our trained models on images w.r.t two aspects: times of day and weather, and find that in spite of facing the same challenges as prior methods, our trained models significantly outperform them.
-developed the STEAM Labs TM program to engage middle and high school students in learning science, technology, engineering, arts, and math concepts through designing and building chain reaction machines. He founded and led teams to two collegiate Rube Goldberg Machine Contest national championships, and has appeared on many TV shows (including Modern Marvels on The History Channel and Jimmy Kimmel Live on ABC) and a movie with his chain reaction machines. He serves on the Board of the i.d.e.a. Museum in Mesa, AZ, and worked as a behind-the scenes engineer for season 3 of the PBS engineering design reality TV show Design Squad. He also held the Guinness World Record for the largest number of steps -125 -in a working Rube Goldberg machine. Dr. Micah Lande, Arizona State UniversityMicah Lande, Ph.D. is an Assistant Professor in the Engineering and Manufacturing Engineering programs and Tooker Professor at the Polytechnic School in the Ira A. Fulton Schools of Engineering at Arizona State University. He teaches human-centered engineering design, design thinking, and design innovation project courses. Dr. Lande researches how technical and non-technical people learn and apply a design process to their work. He is interested in the intersection of designerly epistemic identities and vocational pathways. Dr. Lande is the PI/co-PI on NSF-funded projects focused on engineering doing and making, citizen science and engineering outreach, and "revolutionizing" engineering education. He has also been an instructor and participant in the NSF Innovation Corps for Learning program. He received his B.S in Engineering (Product Design), M.A. in Education (Learning, Design and Technology) and Ph.D. in Mechanical Engineering (Design Education) from Stanford University. Steven Weiner, Arizona State University, Polytechnic campusSteven Weiner is a PhD student in Human and Social Dimensions of Science and Technology at the School for the Future of Innovation in Society at Arizona State University. He is interested in researching innovative learning frameworks at the intersection of formal and informal STEM education, specifically focusing on the impact of long-term, project-based programs on middle and high school students at community makerspaces and science centers. Before starting his doctoral studies, Mr. Weiner served as the founding Program Director for CREATE at Arizona Science Center, a hybrid educational makerspace/ community learning center. He has previous experience as a physics and math instructor at the middle school and high school levels.c American Society for Engineering Education, 2017 Engineering Students Rapidly Learning at Hackathon Events IntroductionHackathon events are severely time constrained events technological invention sprints where programmers and engineers dive deeply into solving design challenges. Part of the experience is to learn as much as you can in a short amount of time, usually 24-36 hours consecutively. A majority of participants enter hackathons not knowing the direction of their intended pro...
No abstract
is a first-year Ph.D. student at Arizona State University (ASU) studying Engineering Education Systems & Design. She has received her M.S./B.S. in Software Engineering through an accelerated program at ASU. She began researching hackathons after she joined the Fulton Undergraduate Research Initiative (FURI) in her junior year. This stemmed from her love of learning in hackathons having participated in numerous hackathons from as far west as Southern California to as far east as Pennsylvania.
No abstract
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.