Learn More
In this paper we evaluate the quality of the activation layers of a convolutional neural network (CNN) for the generation of object proposals. We generate hypotheses in a sliding-window fashion over different activation layers and show that the final convolutional layers can find the object of interest with high recall but poor localization due to the(More)
Object detection is a challenging task in visual understanding domain, and even more so if the supervision is to be weak. Recently, few efforts to handle the task without expensive human annotations is established by promising deep neural network. A new architecture of cascaded networks is proposed to learn a convolutional neural network (CNN) under such(More)
A total of 52 jaundiced elderly patients who had malignant obstruction of the distal common bile duct and who required palliative biliary decompression were randomized to receive either an endoscopically placed biliary endoprosthesis (10 French gauge) or conventional surgical bypass. Patients within the two treatment groups were well matched and 51 were(More)
Users often have very specific visual content in mind that they are searching for. The most natural way to communicate this content to an image search engine is to use key-words that specify various properties or attributes of the content. A naive way of dealing with such multi-attribute queries is the following: train a classifier for each attribute(More)
A patient who had a 50% gastrectomy with a Billroth II gastrojejunostomy one and a half years previously, complained of recent severe weakness as the only symptom was found to have an iron deficiency anemia with a periumbilical mass. A gastrointestinal series showed a soft tissue density in the epigastric area which, by ultrasonography, was found to be(More)
The use of elective tracheostomy in major head and neck surgery is well established, although practice varies between units. There is no published method that reliably predicts the need for tracheostomy. This paper describes the development of a surgical scoring system designed to achieve that aim. The system was devised using data obtained retrospectively(More)
The video and action classification have extremely evolved by deep neural networks specially with two stream CNN using RGB and optical flow as inputs and they present outstanding performance in terms of video analysis. One of the shortcoming of these methods is handling motion information extraction which is done out side of the CNNs and relatively time(More)
The recognition of human actions and the determination of human attributes are two tasks that call for fine-grained classification. Indeed, often rather small and inconspicuous objects and features have to be detected to tell their classes apart. In order to deal with this challenge, we propose a novel convolutional neural network that mines mid-level image(More)