We propose a computational attention approach to localize and classify characters in a sequence in a given image. Our approach combines spatial soft-attention with attention regularization and learns “where-to-look” to ...

Deep learning can cope with complex tasks; however, it suffers from the crucial problem of high dimensionality which makes it intractable to learn the patterns and classify the data. Importantly, having a fixed model cannot ...

We created Echo, a tool designed to help bridge the gap between game analytics and video footage analysis. Echo reconstructs gameplay sessions from recorded data and presents them with their original graphics. A comparative ...

Metal Big Area Additive Manufacturing is an additive manufacturing technique based on gas metal arc welding. The systems’ dual nozzle design prints two steels simultaneously; in this study, the test samples are made from ...

Containers and virtual machines are being adopted to develop embedded Linux Internet-of-Things applications. Consumer Internet-of-Things devices have been notoriously insecure due to loss of continued software support. To ...

We propose a new approach to data driven constraint programming. By extending the relational model to handle constraints and variables as first class citizens, we are able to express first order logic SAT problems using ...

Image registration is an indispensable tool in medical image analysis. Traditionally, registration algorithms are aimed at aligning image pairs using regularizers to impose smoothness restrictions on unknown deformation ...

Digital games are a complex interactive medium providing a multitude of different experiences. The field of games user research (GUR) is dedicated to investigating and optimizing user experience in games. Such inquiries ...

Carbon based thin-film materials have been attracting a lot of attention in many research fields due to their unique material properties. Detection of ionizing radiation is an important application that carbon and diamond ...

Recently, Variational Autoencoders (VAEs) have shown remarkable performance in collaborative filtering (CF) with implicit feedback. These existing recommendation models learn user representations to reconstruct or predict ...