We propose an end-to-end learning framework for foreground object segmentation. Given a single novel image, our approach produces a pixel-level mask for all “object-like” regions—even for object categories never seen during training. We formulate the task as a structured prediction problem of assigning a foreground/background label to each pixel, implemented using a deep fully convolutional net...

Modern chips may contain a large number of embedded test, debugging, configuration, and monitoring features, called instruments. An instrument and its instrument access procedures may be pre-developed and reused, and each instrument—in different chips and through the life-time of a chip—may be accessed in different ways, which requires retargeting. To address reuse and retargeting of instrument...

Several local features have become quite popular for concept detection and search, due to their ability to capture distinctive details. Typically a Bag of Words approach is followed, where a codebook is built by quantizing the local features. In this paper, we propose to represent SIFT local features extracted form an image as a multivariate Gaussian distribution, obtaining a mean vector and a ...

Insertion of a single-chain variable-fragment antibody (scFv) to HER2 (human epidermal growth factor receptor 2) in gD, gH, or gB gives rise to herpes simplex viruses (HSVs) specifically retargeted to HER2-positive cancer cells, hence to highly specific nonattenuated oncolytic agents. Clinical-grade virus production cannot rely on cancer cells. Recently, we developed a double-retargeting strate...

Oncolytic viruses gain cancer specificity in several ways. Like the majority of viruses, they grow better in cancer cells that are defective in mounting the host response to viruses. Often, they are attenuated by deletion or mutation of virulence genes that counteract the host response or are naturally occurring oncolytic mutants. In contrast, retargeted viruses are not attenuated or deleted; t...

This paper introduces a complete methodology for retargeting of transistor-level circuits to different sets of specifications. By careful integration of the device sizing and layout generation tasks, fully functional designs are generated in a few minutes of CPU time. The methodology is illustrated via the retargeting of a fully-differential Miller-compensated two-stage operational amplifier fo...

Tracking the movement of vehicles in urban environments using fixed position sensors, mobile sensors, and crowd-sourced data is a challenging but important problem in applications such as law enforcement and defense. A dynamic data driven application system (DDDAS) is described to track a vehicle’s movements by repeatedly identifying the vehicle under investigation from live image and video dat...

................................................................................................................................... 2 Acknowledgement .................................................................................................................... 3 Table of Content ..................................................................................................................