Inception Institute of Artificial Intelligence, Ltd.

Tiandatz Technology Co., Ltd

Yanwei Pang

Ling Shao

Fahad Shahbaz Khan

Hisham Cholakkal

Rao Muhammad Anwer

Jiale Cao

Yiming Liu

Huawei Xia

Yishun Guo

Disclosed is a method and system for designing a magnetic resonance imaging permanent magnet, a device, and a medium. The method includes: obtaining a target magnetic field map of a target magnetic block set, and the target magnetic block set is a set of target magnetic blocks corresponding to a number of target magnetic blocks determined based on a total mass of a set magnet; inputting the target magnetic field map into a parametric model to obtain a parameterized matrix, where the parameterized matrix includes a candidate spatial position number of each target magnetic block, spatial position coordinates of a center of each target magnetic block, an angle of each target magnetic block, and material distribution of each magnetic block, and the parametric model is constructed based on a deep neural network; and determining a three-dimensional spatial structure and material distribution of a permanent magnet based on the parameterized matrix.

Method and system for designing magnetic resonance imaging permanent magnet, device, and medium

This disclosure relates to improved techniques for performing computer vision functions, including common object detection and instance segmentation. The techniques described herein utilize neural network architectures to perform these functions in various types of images, such as natural images, UAV images, satellite images, and other images. The neural network architecture can include a dense location regression network that performs object localization and segmentation functions, at least in part, by generating offset information for multiple sub-regions of candidate object proposals, and utilizing this dense offset information to derive final predictions for locations of target objects. The neural network architecture also can include a discriminative region-of-interest (RoI) pooling network that performs classification of the localized objects, at least in part, by sampling various sub-regions of candidate proposals and performing adaptive weighting to obtain discriminative features.

Dense and discriminative neural network architectures for improved object detection and instance segmentation

Growing community of inventors

Tianjin, China

Yanwei Pang