Learn More
Visual question answering (VQA) task not only bridges the gap between images and language, but also requires that specific contents within the image are understood as indicated by linguistic context of the question, in order to generate the accurate answers. Thus, it is critical to build an efficient embedding of images and texts. We implement DualNet,(More)
The workflow of extracting features from images using con-volutional neural networks (CNN) and generating captions with recurrent neural networks (RNN) has become a de-facto standard for image captioning task. However, since CNN features are originally designed for classification task, it is mostly concerned with the main conspicuous element of the image,(More)
This paper characterized bovine extraocular muscles (EOMs) using creep, which represents long-term stretching induced by a constant force. After preliminary optimization of testing conditions, 20 fresh EOM samples were subjected to four different loading rates of 1.67, 3.33, 8.33, and 16.67%/s, after which creep was observed for 1,500 s. A published(More)
Recent advances in image captioning task have led to increasing interests in video captioning task. However, most works on video captioning are focused on generating single input of aggregated features, which hardly deviates from image cap-tioning process and does not fully take advantage of dynamic contents present in videos. We attempt to generate video(More)
Visual Question Answering (VQA) task has showcased a new stage of interaction between language and vision, two of the most pivotal components of artificial intelligence. However, it has mostly focused on generating short and repetitive answers, mostly single words, which fall short of rich linguistic capabilities of humans. We introduce Full-Sentence Visual(More)
Developing a system that can automatically respond to a user's utterance has recently become a topic of research in natural language processing. However, most works on the topic take into account only a single preceding utterance to generate a response. Recent works demonstrate that the application of statistical machine translation (SMT) techniques towards(More)
Recently there have been major advances in the electro-mechanical design of upper extremity prosthetics. However, the development of control strategies for such prosthetics has lagged significantly behind. Conventional noninvasive myoelectric control strategies rely on the amplitude of electromyography (EMG) signals from flexor and extensor muscles in the(More)
The Poisson ratio (PR) is a fundamental mechanical parameter that approximates the ratio of relative change in cross sectional area to tensile elongation. However, the PR of extraocular muscle (EOM) is almost never measured because of experimental constraints. The problem was overcome by determining changes in EOM dimensions using computed X-ray tomography(More)
Compliance with, implementation of, and monitoring of evidence-based guidelines can be challenging in clinical environments. We created a real-time patient safety dashboard, deployed since June 2011. We report on our current research to make it very easy to go from a clinical specification to deployment, thus enabling clinicians to specify (and change)(More)
PURPOSE Intramuscular innervation of horizontal rectus extraocular muscles (EOMs) is segregated into superior and inferior (transverse) compartments, while all EOMs are also divided into global (GL) and orbital (OL) layers with scleral and pulley insertions, respectively. We sought evidence of potential independent action by examining passive mechanical(More)
  • 1