-
1.发明公开公开(公告)号：US20240346802A1
公开(公告)日：2024-10-17
申请号：US18632666
申请日：2024-04-11
Applicant: Fujitsu Limited
Inventor： Mengjiao WANG , Rujie LIU
CPC classification number: G06V10/761 , G06T7/248 , G06V10/751 , G06T2207/10016
Abstract: An information processing device, an information processing method and a computer readable storage medium are disclosed. The information processing device includes: processing circuitry configured to calculate, for each tracklet, a similarity set of each frame of a predetermined frame set included in the tracklet; determine, for each tracklet, a splitting point of the tracklet from the predetermined frame set based on the similarity set; split the tracklet into multiple sub-segments by using the determined splitting point; and merge sub-segments which involve a same object and do not overlap temporally among sub-segments to be merged, to obtain a merged segment, where the sub-segments to be merged include the multiple sub-segments obtained by the splitting unit.
-
公开(公告)号：US20190114468A1
公开(公告)日：2019-04-18
申请号：US16159031
申请日：2018-10-12
Applicant: FUJITSU LIMITED
Inventor： Pingping LU , Rujie LIU
Abstract: A method and apparatus for evaluating an illumination condition in a face image is provided by decomposing a face image into illumination feature components and face feature components; extracting determined areas in the face image; calculating a maximum luminance feature, a minimum luminance feature and an illumination direction feature according to the decomposed illumination feature components in the determined areas. The illumination condition in the face image is evaluated according to the maximum luminance feature, the minimum luminance feature and the illumination direction feature.
-
公开(公告)号：US20210097557A1
公开(公告)日：2021-04-01
申请号：US16851759
申请日：2020-04-17
Applicant: FUJITSU LIMITED
Inventor： Rujie LIU
Abstract: A method and an apparatus for optimizing object prediction and a storage medium are provided according to the present disclosure. The method includes: grouping multiple objects, where each group of objects have similar characteristics; building a predictor library for each group of objects, respectively; determining, in the predictor library of each group of objects, an initial corresponding predictor for each object, based on historical characteristic data with a fixed length of time related to each object; and dynamically updating the initial corresponding predictor for each object, respectively, by using characteristic data varying with time and related to each object, where a prediction performance of the updated corresponding predictor is optimal with respect to the characteristic data varying with time and related to each object.
-
公开(公告)号：US20190392248A1
公开(公告)日：2019-12-26
申请号：US16450153
申请日：2019-06-24
Applicant: FUJITSU LIMITED
Inventor： Meng ZHANG , Rujie LIU
Abstract: The present disclosure relates to an information processing method and an information processing apparatus. The information processing method according to the present disclosure performs training on a classification model by using a plurality of training samples, and comprises the steps of: adjusting a distribution of feature vectors of the plurality of training samples in a feature space based on a typical sample in the plurality of training samples; and performing training on the classification model by using the adjusted feature vectors of the plurality of training samples. Through the technology according to the present disclosure, it is possible to perform pre-adjustment on training samples before training, such that it is possible to reduce discrimination between training samples belonging to a same class and increase discrimination between training samples belonging to different classes in the training process. The classification model trained as such is capable of performing accurate classification on samples acquired under an extreme condition.
-
5.发明申请公开(公告)号：US20190286940A1
公开(公告)日：2019-09-19
申请号：US16299177
申请日：2019-03-12
Applicant: FUJITSU LIMITED
Abstract: The method for training an image model, in each round of training performed with respect to each sample image: inputs an image obtained by cropping the sample image by an object extraction component obtained through a previous round of training, as a scale-adjusted sample image, into the image model, wherein the object extraction component is used for extracting concerned objects in sample images at respective scales; inputs a feature of the scale-adjusted sample image into a local classifier in the image model respectively, performs category prediction with respect to feature points in the feature, so as to obtain a local prediction result, and updates the object extraction component based on the local prediction result; performs object level category prediction for the scale-adjusted sample image based on the feature and the updated object extraction component; and trains the image model based on a category prediction result of the scale-adjusted sample image.
-
Patent Agency Ranking
6.发明公开公开(公告)号：US20230196735A1
公开(公告)日：2023-06-22
申请号：US18047780
申请日：2022-10-19
Applicant: Fujitsu Limited
Inventor： Meng ZHANG , Rujie LIU
IPC: G06V10/774 , G06V10/82 , G06V10/776
CPC classification number: G06V10/774 , G06V10/82 , G06V10/776
Abstract: A method and an apparatus for training a neural network, an image recognition method and a computer readable storage medium are disclosed. The neural network includes a first model and a second model. The method for training a neural network includes: acquiring a second image from a first image, wherein a quality of the second image is lower than that of the first image; inputting the first image into the first model of the neural network, and inputting the second image into the second model of the neural network; calculating an attention map and a gradient map of the first model and an attention map and a gradient map of the second model; constructing a loss function based on a matrix of a dot product of the gradient map and the attention map of the first model and a matrix of a dot product of the gradient map and the attention map of the second model; and training the neural network by minimizing the loss function.
-
公开(公告)号：US20150187075A1
公开(公告)日：2015-07-02
申请号：US14584153
申请日：2014-12-29
Applicant: FUJITSU LIMITED
IPC: G06T7/00
CPC classification number: G06T7/0002 , G06T2207/30168 , G06T2207/30176
Abstract: The present invention discloses a bleed-through detection method and a bleed-through detection device. The method includes: obtaining a recto image and a verso image, thereby obtaining pixel pairs including the first points and the corresponding second points; determining some foreground pixels and some background pixels; performing modeling for four types of pixel pairs, so as to form four models; calculating, for a pixel pair that hasn't been modeled, similarities of the pixel pair with respect to the four models respectively, so as to determine a type of the pixel pair; and judging, as bleed-through on the verso image, a second point determined as a background pixel which corresponds to a first point determined as a foreground pixel, and judging, as bleed-through on the recto image, a first point determined as a background pixel which corresponds to a second point determined as a foreground pixel.
Abstract translation: 本发明公开了一种渗漏检测方法和渗透检测装置。 该方法包括：获得直角图像和二维图像，从而获得包括第一点和对应的第二点的像素对; 确定一些前景像素和一些背景像素; 对四种像素对执行建模，从而形成四个模型; 对于尚未建模的像素对，分别计算像素对相对于四个模型的相似度，以便确定像素对的类型; 并且作为所述通配图像上的渗透判断为被确定为与被确定为前景像素的第一点对应的背景像素的第二点，并且将作为所述直角图像上的渗透判定为第一点，被确定为 对应于被确定为前景像素的第二点的背景像素。
-
公开(公告)号：US20200334490A1
公开(公告)日：2020-10-22
申请号：US16745375
申请日：2020-01-17
Applicant: FUJITSU LIMITED
IPC: G06K9/62 , G06N20/00 , G06F16/56 , G06K9/46 , G06F16/532 , G06F16/587
Abstract: The application relates to an image processing apparatus, and a training method and training apparatus for training the image processing apparatus. The training apparatus comprises: a feature map extracting unit to extract feature maps of support images and a query image; a refining unit to determine, with respect to each support image, a matching feature vector, based on the feature maps; and a joint training unit to use a training image as the query image to execute joint training, such that it is capable of determining a matching support image and a matching location with respect to a new query image, the training image matching a specific support image. The image processing apparatus trained through the above training technique is capable of simultaneously determining a matching support image among a plurality of support images respectively belonging to different classes which matches a query image, and determining a matching location.
-
公开(公告)号：US20190385086A1
公开(公告)日：2019-12-19
申请号：US16438664
申请日：2019-06-12
Applicant: FUJITSU LIMITED
Inventor： Mengjiao WANG , Rujie LIU
Abstract: There are provided a method of knowledge transferring, an information processing apparatus and a storage medium. The method of knowledge transferring includes: obtaining a first model which has been trained in advance with respect to a predetermined task; and training a second model with respect to the predetermined task by utilizing a comprehensive loss function, such that the second model has knowledge of the first model. The comprehensive loss function is based on a first loss function weighted by accuracy of an output result of the first model for a training sample in regard to the predetermined task, and a second loss function. The first loss function represents a difference between processing results of the second model and the first model for the training sample. The second loss function represents accuracy of an output result of the second model for the training sample in regard to the predetermined task.
-
公开(公告)号：US20180336438A1
公开(公告)日：2018-11-22
申请号：US15971549
申请日：2018-05-04
Applicant: FUJITSU LIMITED
Inventor： Ziqiang SHI , Liu LIU , Rujie LIU
CPC classification number: G06K9/6277 , G06K9/481 , G06K9/6215 , G06K9/6256 , G06K9/6262 , G10L17/00
Abstract: A multi-view vector processing method and a multi-view vector processing device are provided. A multi-view vector x represents an object containing information on at least two non-discrete views. A model of the multi-view vector, where the model includes at least components of: a population mean μ of the multi-view vector, view component of each view of the multi-view vector and noise is established. The population mean μ, parameters of each view component and parameters of the noise , are obtained by using training data of the multi-view vector x. The device includes a processor and a storage medium storing program codes, and the program codes implements the aforementioned method when being executed by the processor.
-
-
-
-
-
-
-
-
-
Search Results
20
records
(took 0.015 seconds)
