Search

The main intent of multiple shape tracking is to allocate distinct track identities for all the salient shapes in a video sequence. The approach used in this problem is ‘Track by Detection’ scheme, where it needs to first, find shapes in a frame, map shapes to appropriate tracking tracks, and maintain the data continuity model throughout the video sequence. This thesis proposes a process to address shape tracking by maintaining a data continuity model based on shape relative proximity, shape context descriptor, and spatial color variation of a shape. Shape intersection is used to calculate the shape proximity. A spatial color analyzer is used to compare shape’s color variations between two consecutive frames, and shape similarity comparison is done by using the shape context descriptor. These values are then used to build and maintain the data continuity model throughout the video sequence. Besides, this also introduces a methodology to reestablish tracking identity when the detection is failed or occluded temporarily between small numbers of frames. The proposed method can be implemented using both online and offline approaches. To assess the proposed method’s robustness, we evaluated the online method in both VOT-ST2020 (Visual Object Tracking Challenge - Short Term) and MOT20 (Multiple Object Tracking) benchmark datasets. It was observed that in both benchmarks, the proposed approach can deliver state-of-the-art results.

Over the past decade, researchers have utilized novel technologies to improve the lives of the elderly population. Virtual Reality (VR) is among the most promising platforms that could help the elderly stay cognitively active; However, the extent to which this population is willing to seek out and engage in VR activities remains unclear. In this document, we have discussed our two studies regarding VR and seniors. Our first study is a pilot project including three senior residents of Winnipeg. In this study, we assessed the impact of VR game name “DoVille” on the cognitive capacities of the elderly. We designed a two-week procedure with 20 minutes of VR sessions per day. While the comparison of pre-DoVille and post-DoVille test scores were statistically insignificant, we have gained valuable information about the feasibility and possible challenges of similar projects in the future. Among these issues, we have discussed the eligibility criteria, VR sessions’ setting and length of training sessions for the seniors. The second study is a survey project assessing the attitudes of the elderly toward VR during the COVID-19 pandemic. All senior residents of Manitoba between the ages of 65 and 90 were eligible to participate in the study. The survey was administered online and by phone, and 103 individuals responded to our questionnaire. Our results suggest that a large proportion of elderly individuals have become interested in VR technology as a result of the COVID-19 pandemic. We developed two models for VR use based on the responses. Our model of VR use for communication/interaction could account for approximately 50% of the variance in interest levels in VR, and our model of VR use for cognitive benefits accounted for 35% of the variance. These models included variables such as previous experience with technology, age and gender. In conclusion, these two studies provide us with a better understanding of the elderly’s interest in technology and how we could implement n

Alzheimer’s disease (AD) is typically characterized by hypometabolism in medial-frontal lobes, posterior cingulum and temporal lobes which can be assessed using fluorodeoxyglucose (FDG) positron emission tomography (PET) in vivo. Hypermetabolism in cerebellum and pons have been also consistently reported, but it is often neglected. Some studies have dismissed it as an artifact of global mean normalization. We hypothesize that the hypermetabolic regions are also important in disease pathology in AD. Using 88 AD subjects and 88 age-sex matched Normal Controls (NL1) from the publicly available neuroimaging database enabled by Alzheimer’s disease Neuroimaging Initiative (ADNI; http://adni.loni.usc.edu/ ), we established FDG PET-based AD Classifier (FAC) that differentiated AD patients from normal individuals (sensitivity= 87.50%, specificity = 82.95%). To rule out that the observed hypermetabolism is an artifact induced by global signal intensity normalization, we replicated FAC using white matter (WM) as a reference region, and we confirmed the presence of hypermetabolism in cerebellum and pons. The brain metabolic network of AD and NL has been further analyzed using graph theory. The differences in Betweenness Centrality (BC, a measure of hubness of information flow) between AD vs NL network were correlated with region weights of FAC, which was driven by a few brain regions. In particular, the hypermetabolism in cerebellum was accompanied with higher BC while the hypometabolism in temporal lobe and posterior cingulum were accompanied with lower BC in AD. The brain regions with higher BC in AD network showed a progressive increase in FDG uptake (hypermetabolism) over 2 years in prodromal AD patients (n=39), whose baseline scans were acquired at least 2 years prior to the AD diagnosis. This progressive increase of FDG uptake was not observed in the control subjects who had mild cognitive impairment but did not develop AD over the next 2 years (n=39) nor in NL who were f