Výsledky vyhledávání - "Meyer, Benjamin J"

Report

Multimorbidity Content-Based Medical Image Retrieval Using Proxies

Autor: Xing, Yunyan, Meyer, Benjamin J., Harandi, Mehrtash, Drummond, Tom, Ge, Zongyuan

Content-based medical image retrieval is an important diagnostic tool that improves the explainability of computer-aided diagnosis systems and provides decision making support to healthcare professionals. Medical imaging data, such as radiology image

Externí odkaz: http://arxiv.org/abs/2211.12185

Zobrazit plný text záznamu

Report

OpenGAN: Open Set Generative Adversarial Networks

Autor: Ditria, Luke, Meyer, Benjamin J., Drummond, Tom

Many existing conditional Generative Adversarial Networks (cGANs) are limited to conditioning on pre-defined and fixed class-level semantic labels or attributes. We propose an open set GAN architecture (OpenGAN) that is conditioned per-input sample w

Externí odkaz: http://arxiv.org/abs/2003.08074

Zobrazit plný text záznamu

Report

The Importance of Metric Learning for Robotic Vision: Open Set Recognition and Active Learning

Autor: Meyer, Benjamin J., Drummond, Tom

State-of-the-art deep neural network recognition systems are designed for a static and closed world. It is usually assumed that the distribution at test time will be the same as the distribution during training. As a result, classifiers are forced to

Externí odkaz: http://arxiv.org/abs/1902.10363

Zobrazit plný text záznamu

Report

Deep Metric Learning and Image Classification with Nearest Neighbour Gaussian Kernels

Autor: Meyer, Benjamin J., Harwood, Ben, Drummond, Tom

We present a Gaussian kernel loss function and training algorithm for convolutional neural networks that can be directly applied to both distance metric learning and image classification problems. Our method treats all training features from a deep n

Externí odkaz: http://arxiv.org/abs/1705.09780

Zobrazit plný text záznamu