Masataka Yamaguchi

Learn More
The workflow of extracting features from images using convolutional 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)
Generating captions to describe images is a fundamental problem that combines computer vision and natural language processing. Recent works focus on descriptive phrases, such as "a white dog" to explain the visual composites of an input image. The phrases can not only express objects, attributes, events, and their relations but can also reduce visual(More)
We performed positron emission tomography (PET) using 2-deoxy-2-[(18)F]fluoro-D-glucose ([(18)F]FDG) to evaluate the pharmacokinetics of nasal drug absorption in the rat. The dosing solution of [(18)F]FDG was varied in volume (ranging from 5 to 25 μl) and viscosity (using 0% to 3% concentrations of hydroxypropylcellulose). We modeled the pharmacokinetic(More)
In this paper, we address the problem of spatio-temporal person retrieval from multiple videos using a natural language query, in which we output a tube (i.e., a sequence of bounding boxes) which encloses the person described by the query. For this problem, we introduce a novel dataset consisting of videos containing people annotated with bounding boxes for(More)
OBJECTIVE The purpose of this study was to evaluate, under various conditions, the damage that might occur to viable cells after CO2 laser irradiation of the root canal. STUDY DESIGN A laser tip was placed within the root canal of extracted human teeth that were positioned above cultured fibroblasts (NB cells), and then irradiation was applied. The(More)
  • 1