白丝美女被狂躁免费视频网站,500av导航大全精品,yw.193.cnc爆乳尤物未满,97se亚洲综合色区,аⅴ天堂中文在线网官网

Systems and methods for 3D image distification

專利號
US11176414B1
公開日期
2021-11-16
申請人
STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY（US IL Bloomington）
發(fā)明人
Elizabeth Flowers; Puneit Dua; Eric Balota; Shanna L. Phillips
IPC分類
G06K9/62; G06K9/42; G06K9/00
技術(shù)領(lǐng)域
3d,2d,image,images,or,computing,matrix,in,2d3d,model
地域： IL IL Bloomington

摘要

Systems and methods are described for Distification of 3D imagery. A computing device may obtain a three dimensional (3D) image that includes rules defining a 3D point cloud used to generate a two dimensional (2D) image matrix. The 2D image matrix may include 2D matrix point(s) mapped to the 3D image, where each 2D matrix point can be associated with a horizontal coordinate and a vertical coordinate. The computing device can generate an output feature vector that includes, for at least one of the 2D matrix points, the horizontal coordinate and the vertical coordinate of the 2D matrix point, and a depth coordinate of a 3D point in the 3D point cloud of the 3D image. The 3D point can have a nearest horizontal and vertical coordinate pair that corresponds to the horizontal and vertical coordinates of the at least one 2D matrix point.

說明書

In various embodiments, for example, a 2D image or a 3D image prediction model may use a convolutional neural network (“ConvNet” or “CNN”) model to classify image behaviors. CNNs are a machine learning type of predictive model that can be used for image recognition and classification. CNNs can operate on 2D or 3D images, where, for example, such images are represented as a matrix of pixel values. In certain embodiments, a Distification method may be used with a CNN model to predict driver behavior and/or gestures for 3D images.

Generally, a CNN can be used to determine one or more classifications for a given image by passing the image through a series of computational operational layers, as described herein. By training and utilizing theses various layers, a CNN model can determine a probability that an image belongs to a particular class.

For example, for the image 650 of FIG. 6B, the classifications and probabilities may be “normal driving” (20%) and “texting” (50%) as indicated by points 660 and 662, respectively, because, while the driver's hands are on the steering wheel (point 660) in the image 650 (which can increase the probability for “normal driving” classification), the use of the mobile phone (point 662) can increase the probability for the “texting” classification. In some embodiments, the identification of “texting” (or other negative driving behaviors) may be heavier weighted in the CNN model, such that an identification of “texting,” etc., can increase the probability associated with the “texting” classification more than the identification of a “normal driving” classification.

權(quán)利要求

1
微信群二維碼
意見反饋