International Business Machines Corporation

Chinese Academy of Sciences

Xiu Li Li

Xin Zhang

Peng Gao

Xiaolu Zhang

Shi Lei Zhang

Yong Qin

Shi Wan Zhao

Qingyun Di

Shiwan Zhao

Yu Wang

Guo Tong Xie

Guoqiang Xue

Huaifeng Sun

Dongyang Hou

Dingli Gao

Techniques facilitating attention based sequential image processing are provided. A system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise an initialization component that can perform self-attention based training on a model that comprises context information associated with a sequence of images. Images of the sequence of images can be selected during the self-attention based training. The computer executable components can also comprise a localization component that can extract local information from the images selected during the self-attention based training based on the context information. In addition, the computer executable components can also comprise an integration component that can update the model based on an end-to-end integrated attention training framework comprising the context information and the local information.

Attention based sequential image processing

Methods, systems, and storage components for utilizing a deep neural network(s) for classifying at least one medical image. A deep neural network (DNN) can be configured by an image processing component to contain at least one revived neuron, where the revived neuron has an adjusted value based on a focus-learning function that is transferred to the DNN by the image processing component, where the focus-learning function provides the adjustment by updating the DNN with data that contains a corrected classification with respect to at least one normal image sample derived from a medical image, and where the correction is based on the focus-learning function comparing an annotation associated with an abnormal image sample derived from the medical image to another annotation associated with the at least one normal image.

Deep image classification of medical images

The present invention relates to a correction method for transient electromagnetic method-based prediction of water-bearing structure ahead of tunnel wall, which pertains to the technical field of transient electromagnetic sounding method. The correction method includes a method for eliminating effects of transition process in a reception probe and a method of using earth conductivity to eliminate effects of transition process. By using the correction technique, more accurate information about the location, size and shape of the object in front of the tunnel wall can be obtained, which is of great significance to the study of the fine water-bearing geological structure.

Correction method for transient electromagnetic method-based prediction of water-bearing structure ahead of tunnel wall

A method and system of detecting abnormalities captured in a medical image are provided. A medical image is received having one or more lesions and one or more interferences. One or more convolution layers are applied to the medical image to create one or more feature maps, respectively. A region proposal network (RPN) is applied to each of the one or more feature maps to create a bounding box around each lesion and interference. For each bounding box, an object in the bounding box is classified as a lesion, an interference, or a background. Each object that is incorrectly classified as a lesion but determined to be an interference, is stored as a hard-negative example to be part of a training set for a next application of the RPN.

Negative sample enhanced object detection machine

Laptop bag

Growing community of inventors

Beijing, China

Xiu Li Li