7903 research outputs found
Sort by
Exploring Novel Methods for Real-Time Multi-Camera People Tracking in Machine Learning
No full textMulti-camera people tracking is the process of tracking persons and their paths, continuously across different camera fields of view. It can help track suspects across large areas, and assist individuals in emergency situations. The current state-of-the-art (SOTA) method involves using geometric-consistent constraints, information on the appearance of subjects, and pose estimation for dealing with occlusion issues. This current SOTA works well, but is still lacking in its ability to handle occlusion and perform well in real-time applications on a network. With occlusion, the IDs assigned to persons can be accidentally swapped in high density areas, or places where there are drastically different camera angles. In the case of real-time deployment of these methods, the models can suffer from the fact that they are not yet well optimized, and their performance and accuracy can be diminished at the cost of output quality. Using pose estimation models like RTMPose or HRNet could aid in reducing occlusion issues, and using lightweight models that are better suited for scaling while simultaneously maintaining accuracy, like OSNet, can help address performance issues and reduce compute requirements
Litter-acy
No full textThis project for Santa Clara University School of Engineering’s annual Senior Design, Litter-acy, addresses the critical societal issue of improper waste management and the lack of public knowledge regarding proper sorting techniques. These contribute significantly to environmental pollution and deterioration. Through recognizing how traditional educational methods often fail to engage younger audiences and combat the ”adolescent dip” in sustainability interest, Litter-acy proposes an innovative solution: an iOS mobile application that gamifies waste management education.
The app features multiple interactive mini-games focused on sustainable waste sorting, including Connections, Catcher, and Trivia, designed to make learning engaging and accessible. It incorporates key features such as an extensive item bank, detailed tutorials for each game, and provides educational feedback during gameplay. To enhance user engagement and retention, the app includes social features such as global and friend leaderboards for social competition and an adding friends system for community building among users. External educational resources like articles and videos are also built in and available within the app. The project also involved user research and testing, coupled with adherence to Apple’s Human Interface Guidelines, to inform design plans and allow focus on usability and accessibility.
Litter-acy aims to educate and inspire real-life changes in users’ daily routines, empower communities, and contribute to a sustainable future by transforming waste management learning into a fun and motivating activity
Road Object Detection in Fish-Eye Cameras
No full textObject detection plays a crucial role in traffic surveillance for road safety and traffic management. Road Object Detection can be used for monitoring traffic flow or traffic analysis. In traffic monitoring systems, fish-eye cameras are particularly useful because they can cover larger areas of streets and intersections, reducing the need for multiple cameras. Their ability to provide wide, omnidirectional coverage, which traditional cameras with limited fields of view (FoV) cannot offer, is the reason for fish-eye camera’s recent popularity.
However, these cameras also introduce image distortion, requiring complex techniques for undistortion and unwarping, or specialized processing methods to manage the distortion effectively. The AI City Challenge 2024, Track 4 introduces a novel fish-eye camera dataset for the 2D road object detection task, FishEye8K.
Many previous methods rely on ensembles with different combinations of YOLO and transformer models under Weighted Box Fusion (WBF). These techniques are also coupled with image enhancement and super-resolution models to handle images taken at night and low-resolution images respectively. In addition to the image distortion, there is a lack of open fish-eye image datasets for road object detection, with. To combat this, previous techniques generate data through augmenting the VisDrone dataset to be create synthetic fish-eye images. Despite these complex methods, the model performance of the previous top team, VNPT AI, results in an F1-score of 0.6406.
This paper presents a detection framework leveraging RTMDet with an ensemble of predictions to improve robustness against fisheye distortions. We train this new model on both the FishEye8K and augmented VisDrone datasets and incorporate the popular WBF ensemble method. Our combined approach achieves notable results, with an F1-score of 0.6413
SIDEQUE$T
No full textIn recent years, the gap between traditional education and the current labor market has widened dramatically: in 2024, 11. 5 % of California’s 16-24 year olds were neither in school nor employed, alongside a net decline of roughly 3 million students nationwide in the last decade and more than 4 million members of Generation Z left the job despite having conventional degrees. These trends underscore a growing disconnect between academic credentials and market-ready skills and a pressing need for alternative pathways to gainful work. We introduce SIDEQUET’s recommendation engine delivers highly tailored task and gig opportunities—no formal degree required. Our platform vets and categorizes postings, matches them to each user’s skill profile, and continually refines recommendations via feedback and performance data. This paper presents SIDEQUET offers a scalable solution to today’s “worthless degrees” dilemma and paves the way for a more inclusive, skills-driven economy
D.A.M.P.E.R.: Delay and Attenuation Modules for Propagation Emulation and Response
No full textThe DAMPER project, which stands for Delay and Attenuation Modules for Propagation Emulation and Response, is a hands-on system designed to help test how radio signals behave in the real world. When radio waves travel over long distances or through different environments, they often get weaker (attenuated) or arrive later than expected (delayed). DAMPER lets students, researchers, and engineers simulate those conditions in a lab using simple, low-voltage electronics
Seeing Google at 25
No full textSeeing Google at 25
1. History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
A. Fits and starts to dominating the world . . . . . 5
2. Google in the Eyes of Scholars: Google Services 8
A. Digital library . . . . . . . . . . . . . . . . . . . . . . . . . 8
B. Google Earth and Google Maps . . . . . . . . . . . 9
C. Google Fiber . . . . . . . . . . . . . . . . . . . . . . . . . 9
D. Google Glass . . . . . . . . . . . . . . . . . . . . . . . . . 10
E. Google Hangouts and Google+ . . . . . . . . . . . 11
F. Google Lens and Google Clips . . . . . . . . . . . . 11
G. Google Phone (Android) . . . . . . . . . . . . . . . . 12
H. Google Scholar . . . . . . . . . . . . . . . . . . . . . . . . 12
I. Translate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
J. YouTube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3. Google in the Eyes of Scholars: Themes . . . . . . 14
A. Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
B. Ethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
C. Government interactions . . . . . . . . . . . . . . . . 16
D. Explorations and critiques . . . . . . . . . . . . . . . 17
E. Google’s impact on advertising studies . . . . . 17
F. Google’s impact on health
communication studies . . . . . . . . . . . . . . . . . . 18
G. Google’s impact on journalism
and journalism studies . . . . . . . . . . . . . . . . . . 18
H. Google’s impact on pedagogy . . . . . . . . . . . . 19
I. Google’s impact on political communication . 19
J. Voice assistants and communication research . 20
4. Google as a Tool for Scholars . . . . . . . . . . . . . . . 21
A. Evaluations . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
B. N-grams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
C. Environmental communication . . . . . . . . . . . 22
D. Gender and sexuality . . . . . . . . . . . . . . . . . . . 22
E. Other interesting uses . . . . . . . . . . . . . . . . . . . 22
5. Conclusion: Dreams and Fears . . . . . . . . . . . . . . 23
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Book Reviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Party Convergence and Divergence Among Republican Women
No full textRecent work on party asymmetry in the United States characterizes the Democratic Party as a group-oriented party and the Republican Party as an ideologically oriented party. Gender studies of opinion preferences support a group-based conceptualization of the Democratic Party, with women being a represented group, but suggest the possibility that the Republican Party may exhibit greater ideological heterogeneity when differences based upon gender are considered. To this end, we investigate variations in policy opinions using a difference in means analysis comparing women and men congressional donors and men and women non-donating self-identifiers in the Republican Party. We also model congressional contributing among Republican women. Together, the results suggest that Republican women donors’ preferences converge with those of men. However, women non-donors are more moderate than women congressional donors, as well as men congressional donors and non-donors, suggesting there is greater ideological heterogeneity within the Republican Party than studies of party asymmetry report
“Fight for the Education You Deserve”: The Role of Memorable Messages in the Transformation of Navigational Capital for Latinx Community College Adult Learners’ Academic Retention
No full textThis study explores the educational experiences and retention of Latinx California community college adult learners, emphasizing the interplay of cultural, institutional, and individual factors shaping academic persistence. Through a qualitative research study, accompanied by digital collages and an optional post-interview questionnaire, eleven Latinx California community college adult learners across the state engaged with the following research questions: (a) What types of memorable messages do Latinx adult California community college students receive from institutional agents (e.g., counselors and professors) about navigating higher education? (b) How do students transform these memorable messages into navigational capital that supports academic retention? (c) How do students perceive the impact of these messages on their behavior, attitude, and motivation toward their educational goals? Guided by LatCrit, Community Cultural Wealth (CCW), and Anticipatory Socialization through Memorable Messages frameworks, the study examines how these learners navigate institutional structures, utilize resources, and respond to messages from institutional agents (e.g., professors and counselors). Findings reveal that messages communicated by institutional agents impact Latinx adult learners’ academic retention by shaping their self-perception, motivation, and aspirations, with supportive, empathetic messages fostering resilience and counteracting institutional barriers, while institutional neglect undermines students’ academic abilities and sense of belonging. The implications showcase retention strategies for Latinx adult learners, highlighting the need for inclusive, culturally relevant practices, flexible support services, realistic degree completion messaging, diverse leadership, and strategies such as empathetic interactions, enhanced navigational support, and mentorship to improve student persistence