Automated Discovery And Interpretation Of Ada-Compliant Door Placards,
2021
Kutztown University of Pennsylvania
Automated Discovery And Interpretation Of Ada-Compliant Door Placards, John J. Feilmeier
Computer Science and Information Technology Faculty
A familiar difficulty to any new student on campus is making one’s way from classroom A to classroom B. Facilities with different wings, multiple floors, and irregular floorplans can magnify this challenge, while students with vision impairments are impacted even more by the challenge of identifying the destination. This thesis explored different methods of discovering Americans with Disabilities Act (ADA)- compliant room identifying placards (“plaques”) and identifying the text on the sign. The plaque detection was accomplished with both standard image manipulation techniques and a Histogram of Oriented Gradients (HOG) (Dalal & Triggs, 2005) object detector. The text reading utilized …
Contrastive Learning For Unsupervised Auditory Texture Models,
2021
University of Arkansas, Fayetteville
Contrastive Learning For Unsupervised Auditory Texture Models, Christina Trexler
Computer Science and Computer Engineering Undergraduate Honors Theses
Sounds with a high level of stationarity, also known as sound textures, have perceptually relevant features which can be captured by stimulus-computable models. This makes texture-like sounds, such as those made by rain, wind, and fire, an appealing test case for understanding the underlying mechanisms of auditory recognition. Previous auditory texture models typically measured statistics from auditory filter bank representations, and the statistics they used were somewhat ad-hoc, hand-engineered through a process of trial and error. Here, we investigate whether a better auditory texture representation can be obtained via contrastive learning, taking advantage of the stationarity of auditory textures to …
Self-Supervised Learning Disentangled Group Representation As Feature,
2021
Singapore Management University
Self-Supervised Learning Disentangled Group Representation As Feature, Tan Wang, Zhongqi Yue, Jianqiang Huang, Qianru Sun, Hanwang Zhang
Research Collection School Of Computing and Information Systems
A good visual representation is an inference map from observations (images) to features (vectors) that faithfully reflects the hidden modularized generative factors (semantics). In this paper, we formulate the notion of “good” representation from a group-theoretic view using Higgins’ definition of disentangled representation [38], and show that existing Self-Supervised Learning (SSL) only disentangles simple augmentation features such as rotation and colorization, thus unable to modularize the remaining semantics. To break the limitation, we propose an iterative SSL algorithm: Iterative Partition-based Invariant Risk Minimization (IP-IRM), which successfully grounds the abstract semantics and the group acting on them into concrete contrastive learning. …
Acceleration Skinning: Kinematics-Driven Cartoon Effects For Articulated Characters,
2021
Clemson University
Acceleration Skinning: Kinematics-Driven Cartoon Effects For Articulated Characters, Niranjan Kalyanasundaram
All Theses
Secondary effects are key to adding fluidity and style to animation. This thesis introduces the idea of “Acceleration Skinning” following a recent well-received technique, Velocity Skinning, to automatically create secondary motion in character animation by modifying the standard pipeline for skeletal rig skinning. These effects, which animators may refer to as squash and stretch or drag, attempt to create an illusion of inertia. In this thesis, I extend the Velocity Skinning technique to include acceleration for creating a wider gamut of cartoon effects. I explore three new deformers that make use of this Acceleration Skinning framework: followthrough, centripetal stretch, and …
Vireo @ Trecvid 2021 Ad-Hoc Video Search,
2021
Singapore Management University
Vireo @ Trecvid 2021 Ad-Hoc Video Search, Jiaxin Wu, Phuong Anh Nguyen, Chong-Wah Ngo
Research Collection School Of Computing and Information Systems
In this paper, we summarize our submitted runs and results for Ad-hoc Video Search (AVS) task at TRECVid 2020
Component Damage Source Identification For Critical Infrastructure Systems,
2021
University of Arkansas, Fayetteville
Component Damage Source Identification For Critical Infrastructure Systems, Nathan Davis
Graduate Theses and Dissertations
Cyber-Physical Systems (CPS) are becoming increasingly prevalent for both Critical Infrastructure and the Industry 4.0 initiative. Bad values within components of the software portion of CPS, or the computer systems, have the potential to cause major damage if left unchecked, and so detection and locating of where these occur is vital. We further define features of these computer systems and create a use-based system topology. We then introduce a function to monitor system integrity and the presence of bad values as well as an algorithm to locate them. We then show an improved version, taking advantage of several system properties …
A Theory-Driven Self-Labeling Refinement Method For Contrastive Representation Learning,
2021
Singapore Management University
A Theory-Driven Self-Labeling Refinement Method For Contrastive Representation Learning, Pan Zhou, Caiming Xiong, Xiao-Tong Yuan
Research Collection School Of Computing and Information Systems
For an image query, unsupervised contrastive learning labels crops of the same image as positives, and other image crops as negatives. Although intuitive, such a native label assignment strategy cannot reveal the underlying semantic similarity between a query and its positives and negatives, and impairs performance, since some negatives are semantically similar to the query or even share the same semantic class as the query. In this work, we first prove that for contrastive learning, inaccurate label assignment heavily impairs its generalization for semantic instance discrimination, while accurate labels benefit its generalization. Inspired by this theory, we propose a novel …
Fine-Grained Generalization Analysis Of Inductive Matrix Completion,
2021
Singapore Management University
Fine-Grained Generalization Analysis Of Inductive Matrix Completion, Antoine Ledent, Rodrigo Alves, Yunwen Lei, Marius Kloft
Research Collection School Of Computing and Information Systems
In this paper, we bridge the gap between the state-of-the-art theoretical results for matrix completion with the nuclear norm and their equivalent in \textit{inductive matrix completion}: (1) In the distribution-free setting, we prove bounds improving the previously best scaling of \widetilde{O}(rd2) to \widetilde{O}(d3/2√r), where d is the dimension of the side information and rr is the rank. (2) We introduce the (smoothed) \textit{adjusted trace-norm minimization} strategy, an inductive analogue of the weighted trace norm, for which we show guarantees of the order \widetilde{O}(dr) under arbitrary sampling. In the inductive case, a similar rate was previously achieved only under uniform sampling …
Feel And Touch: A Haptic Mobile Game To Assess Tactile Processing,
2021
Center for Scientific Research and Higher Education of Ensenada (CICESE)
Feel And Touch: A Haptic Mobile Game To Assess Tactile Processing, Ivonne Monarca, Monica Tentori, Franceli L. Cibrian
Engineering Faculty Articles and Research
Haptic interfaces have great potential for assessing the tactile processing of children with Autism Spectrum Disorder (ASD), an area that has been under-explored due to the lack of tools to assess it. Until now, haptic interfaces for children have mostly been used as a teaching or therapeutic tool, so there are still open questions about how they could be used to assess tactile processing of children with ASD. This article presents the design process that led to the development of Feel and Touch, a mobile game augmented with vibrotactile stimuli to assess tactile processing. Our feasibility evaluation, with 5 children …
Let's Read: Designing A Smart Display Application To Support Codas When Learning Spoken Language,
2021
Chapman University
Let's Read: Designing A Smart Display Application To Support Codas When Learning Spoken Language, Katie Rodeghiero, Yingying Yuki Chen, Annika M. Hettmann, Franceli L. Cibrian
Engineering Faculty Articles and Research
Hearing children of Deaf adults (CODAs) face many challenges including having difficulty learning spoken languages, experiencing social judgment, and encountering greater responsibilities at home. In this paper, we present a proposal for a smart display application called Let's Read that aims to support CODAs when learning spoken language. We conducted a qualitative analysis using online community content in English to develop the first version of the prototype. Then, we conducted a heuristic evaluation to improve the proposed prototype. As future work, we plan to use this prototype to conduct participatory design sessions with Deaf adults and CODAs to evaluate the …
Facilitating Heuristic Evaluation For Novice Evaluators,
2021
DePaul University
Facilitating Heuristic Evaluation For Novice Evaluators, Anas Abulfaraj
College of Computing and Digital Media Dissertations
Heuristic evaluation (HE) is one of the most widely used usability evaluation methods. The reason for its popularity is that it is a discount method, meaning that it does not require substantial time or resources, and it is simple, as evaluators can evaluate a system guided by a set of usability heuristics. Despite its simplicity, a major problem with HE is that there is a significant gap in the quality of results produced by expert and novice evaluators. This gap has made some scholars question the usefulness of the method as they claim that the evaluation results are a product …
Towards Balancing Vr Immersion And Bystander Awareness,
2021
Singapore Management University
Towards Balancing Vr Immersion And Bystander Awareness, Yoshiki Kudo, Anthony Tang, Kazuyuki Fujita, Isamu Endo, Kazuki Takashima, Yoshifumi Kitamura
Research Collection School Of Computing and Information Systems
Head-mounted displays (HMDs) increase immersion into virtual worlds. The problem is that this limits headset users' awareness of bystanders: headset users cannot attend to bystanders' presence and activities. We call this the HMD boundary. We explore how to make the HMD boundary permeable by comparing different ways of providing informal awareness cues to the headset user about bystanders. We adapted and implemented three visualization techniques (Avatar View, Radar and Presence++) that share bystanders' location and orientation with headset users. We conducted a hybrid user and simulation study with three different types of VR content (high, medium, low interactivity) with twenty …
Intercept Graph: An Interactive Radial Visualization For Comparison Of State Changes,
2021
Singapore Management University
Intercept Graph: An Interactive Radial Visualization For Comparison Of State Changes, Shaolun Ruan, Yong Wang, Qiang Guan
Research Collection School Of Computing and Information Systems
State change comparison of multiple data items is often necessary in multiple application domains, such as medical science, financial engineering, sociology, biological science, etc. Slope graphs and grouped bar charts have been widely used to show a “before-and-after” story of different data states and indicate their changes. However, they visualize state changes as either slope or difference of bars, which has been proved less effective for quantitative comparison. Also, both visual designs suffer from visual clutter issues with an increasing number of data items. In this paper, we propose Intercept Graph, a novel visual design to facilitate effective interactive comparison …
Self-Supervised Multi-Class Pre-Training For Unsupervised Anomaly Detection And Segmentation In Medical Images,
2021
Singapore Management University
Self-Supervised Multi-Class Pre-Training For Unsupervised Anomaly Detection And Segmentation In Medical Images, Yu Tian, Fengbei Liu, Guansong Pang, Yuanhong Chen, Yuyuan Liu, Johan W. Verjans, Rajvinder Singh
Research Collection School Of Computing and Information Systems
Unsupervised anomaly detection (UAD) that requires only normal (healthy) training images is an important tool for enabling the development of medical image analysis (MIA) applications, such as disease screening, since it is often difficult to collect and annotate abnormal (or disease) images in MIA. However, heavily relying on the normal images may cause the model training to overfit the normal class. Self-supervised pre-training is an effective solution to this problem. Unfortunately, current self-supervision methods adapted from computer vision are sub-optimal for MIA applications because they do not explore MIA domain knowledge for designing the pretext tasks or the training process. …
Acoustic/Gravity Wave Phenomena In Wide-Field Imaging: From Data Analysis To A Modeling Framework For Observability In The Mlt Region And Beyond,
2021
Embry-Riddle Aeronautical University
Acoustic/Gravity Wave Phenomena In Wide-Field Imaging: From Data Analysis To A Modeling Framework For Observability In The Mlt Region And Beyond, Jaime Aguilar Guerrero
Doctoral Dissertations and Master's Theses
Acoustic waves, gravity waves, and larger-scale tidal and planetary waves are significant drivers of the atmosphere’s dynamics and of the local and global circulation that have direct and indirect impacts on our weather and climate. Their measurements and characterization are fundamental challenges in Aeronomy that require a wide range of instrumentation with distinct operational principles. Most measurements share the common features of integrating optical emissions or effects on radio waves through deep layers of the atmosphere. The geometry of these integrations create line-of-sight effects that must be understood, described, and accounted for to properly present the measured data in traditional …
Exploiting Reasoning Chains For Multi-Hop Science Question Answering,
2021
Chinese University of Hong Kong
Exploiting Reasoning Chains For Multi-Hop Science Question Answering, Weiwen Xu, Yang Deng, Huihui Zhang, Deng Cai, Wai Lam
Research Collection School Of Computing and Information Systems
We propose a novel Chain Guided Retriever reader (CGR) framework to model the reasoning chain for multi-hop Science Question Answering. Our framework is capable of performing explainable reasoning without the need of any corpus-specific annotations, such as the ground-truth reasoning chain, or human annotated entity mentions. Specifically, we first generate reasoning chains from a semantic graph constructed by Abstract Meaning Representation of retrieved evidence facts. A Chain-aware loss, concerning both local and global chain information, is also designed to enable the generated chains to serve as distant supervision signals for training the retriever, where reinforcement learning is also adopted to …
Mapping E-Commerce Locally And Beyond: Citt K12 Special Investigation Project,
2021
California State University, Long Beach
Mapping E-Commerce Locally And Beyond: Citt K12 Special Investigation Project, Thomas O’Brien, Deanna Matsumoto
Mineta Transportation Institute
As all aspects of the American workplace become automated or digitally enhanced to some degree, K12 educators have an increasing responsibility to help their students acquire the technical skills necessary to organize and interpret information. Increasingly, this is done through Geographic Information Systems (GIS), especially in careers related to transportation and logistics. The Center for International Trade & Transportation (CITT) at CSU Long Beach has developed this K12 Special Investigation Project to introduce ArcGIS StoryMaps, an engaging, accessible and sophisticated web-based GIS application. The lessons center on e-commerce and its accompanying environmental and economic impact. Still, the activities can be …
Wav-Bert: Cooperative Acoustic And Linguistic Representation Learning For Low-Resource Speech Recognition,
2021
Singapore Management University
Wav-Bert: Cooperative Acoustic And Linguistic Representation Learning For Low-Resource Speech Recognition, Guolin Zheng, Yubei Xiao, Ke Gong, Pan Zhou, Xiaodan Liang, Liang Lin
Research Collection School Of Computing and Information Systems
Unifying acoustic and linguistic representation learning has become increasingly crucial to transfer the knowledge learned on the abundance of high-resource language data for low-resource speech recognition. Existing approaches simply cascade pre-trained acoustic and language models to learn the transfer from speech to text. However, how to solve the representation discrepancy of speech and text is unexplored, which hinders the utilization of acoustic and linguistic information. Moreover, previous works simply replace the embedding layer of the pre-trained language model with the acoustic features, which may cause the catastrophic forgetting problem. In this work, we introduce Wav-BERT, a cooperative acoustic and linguistic …
Learning To Teach And Learn For Semi-Supervised Few-Shot Image Classification,
2021
Singapore Management University
Learning To Teach And Learn For Semi-Supervised Few-Shot Image Classification, Xinzhe Li, Jianqiang Huang, Yaoyao Liu, Qin Zhou, Shibao Zheng, Bernt Schiele, Qianru Sun
Research Collection School Of Computing and Information Systems
This paper presents a novel semi-supervised few-shot image classification method named Learning to Teach and Learn (LTTL) to effectively leverage unlabeled samples in small-data regimes. Our method is based on self-training, which assigns pseudo labels to unlabeled data. However, the conventional pseudo-labeling operation heavily relies on the initial model trained by using a handful of labeled data and may produce many noisy labeled samples. We propose to solve the problem with three steps: firstly, cherry-picking searches valuable samples from pseudo-labeled data by using a soft weighting network; and then, cross-teaching allows the classifiers to teach mutually for rejecting more noisy …
Masked Face Analysis Via Multi-Task Deep Learning,
2021
University of Dayton
Masked Face Analysis Via Multi-Task Deep Learning, Vatsa S. Patel, Zhongliang Nie, Trung-Nghia Le, Tam Van Nguyen
Computer Science Faculty Publications
Face recognition with wearable items has been a challenging task in computer vision and involves the problem of identifying humans wearing a face mask. Masked face analysis via multi-task learning could effectively improve performance in many fields of face analysis. In this paper, we propose a unified framework for predicting the age, gender, and emotions of people wearing face masks. We first construct FGNET-MASK, a masked face dataset for the problem. Then, we propose a multi-task deep learning model to tackle the problem. In particular, the multi-task deep learning model takes the data as inputs and shares their weight to …
