Unlock instant, AI-driven research and patent intelligence for your innovation.

Object detection model training and prediction method and device, equipment and medium

A technology for object detection and training methods, applied in the computer field, can solve the problems of expensive lidar, large application limitations, slow speed, etc.

Pending Publication Date: 2020-10-20
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] The existing technology mainly has the following two methods in the field of two-dimensional and three-dimensional detection: the three-dimensional object detection method based on the laser radar point cloud and the three-dimensional object detection method based on the monocular image; among them, the three-dimensional object detection method based on the laser radar point cloud , requires a relatively expensive lidar, and the collected point cloud has sparseness and density inconsistency, which has large application limitations and high cost in actual scenarios
In addition, the 3D object detection method based on monocular images obtains depth through monocular depth estimation, converts the image into a pseudo point cloud, and then applies the method of 3D point cloud detection, but this method requires a depth estimation network and a 2D pre-detection network. It is connected in series with the 3D point cloud detection network, and the speed is relatively slow

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Object detection model training and prediction method and device, equipment and medium
  • Object detection model training and prediction method and device, equipment and medium
  • Object detection model training and prediction method and device, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] figure 1 It is a schematic flow chart of the method for training an object detection model provided in Embodiment 1 of the present application. The method can be executed by a training device or electronic device for an object detection model, and the device or electronic device can be implemented by software and / or hardware. This device or electronic equipment can be integrated in any intelligent equipment with network communication function. Such as figure 1 As shown, the training method of the object detection model may include the following steps:

[0036] S101. When the object detection model to be trained does not satisfy the preset convergence condition, input the current sample image to the object detection model to be trained; perform binary detection on at least one detection object in the current sample image through the object detection model to be trained. Two-dimensional detection, to obtain the two-dimensional prediction image features of each detection...

Embodiment 2

[0045] figure 2 It is a schematic flowchart of the method for training an object detection model provided in Embodiment 2 of the present application. Such as figure 2 As shown, the training method of the object detection model may include the following steps:

[0046] S201. When the object detection model to be trained does not meet the preset convergence condition, input the current sample image to the object detection model to be trained; perform binary detection on at least one detection object in the current sample image through the object detection model to be trained. Two-dimensional detection, to obtain the two-dimensional prediction image features of each detection object and the prediction parameters of the corresponding two-dimensional detection frame.

[0047] In a specific embodiment of the present application, when the object detection model to be trained does not meet the preset convergence conditions, the electronic device can input the current sample image ...

Embodiment 3

[0061] image 3 It is a schematic flowchart of the method for predicting an object detection model provided in Embodiment 3 of the present application. The method can be executed by a prediction device or electronic device for an object detection model, and the device or electronic device can be implemented by software and / or hardware. This device or electronic equipment can be integrated in any intelligent equipment with network communication function. Such as image 3 As shown, the prediction method of the object detection model may include the following steps:

[0062] S301. Input the image to be detected into the pre-trained object detection model; perform two-dimensional detection on at least one detection object in the image to be detected through the pre-trained object detection model, and obtain the two-dimensional predicted image features and corresponding Prediction parameters for 2D detection boxes.

[0063] In a specific embodiment of the present application, th...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an object detection model training and prediction method and device, equipment and a medium, and relates to the field of deep learning and computer vision in artificial intelligence. According to the specific scheme, the method comprises the steps of inputting a current sample image into a to-be-trained object detection model to obtain two-dimensional prediction image features of each detection object and prediction parameters of a two-dimensional detection frame corresponding to the two-dimensional prediction image features; based on the two-dimensional prediction image features of each detection object and the prediction parameters of the corresponding two-dimensional detection frame, obtaining three-dimensional prediction image features of each detection object and prediction parameters of the corresponding three-dimensional detection frame; and training the object detection model according to the two-dimensional prediction image features and the prediction parameters of the two-dimensional detection frames corresponding to the two-dimensional prediction image features, and the three-dimensional prediction image features and the prediction parameters of the three-dimensional detection frames corresponding to the three-dimensional prediction image features. According to the embodiment of the invention, real-time two-dimensional and three-dimensional joint detection can be realized, so that the purposes of improving the detection speed and reducing the detection cost can be achieved.

Description

technical field [0001] The present application relates to the field of computer technology, and further relates to the field of deep learning and computer vision in artificial intelligence, especially a method, device, equipment and medium for training and predicting an object detection model. Background technique [0002] The existing technology mainly has the following two methods in the field of two-dimensional and three-dimensional detection: the three-dimensional object detection method based on the laser radar point cloud and the three-dimensional object detection method based on the monocular image; among them, the three-dimensional object detection method based on the laser radar point cloud , requires a relatively expensive lidar, and the collected point cloud has sparseness and density inconsistency, so the application limitations in actual scenarios are large and the cost is high. In addition, the 3D object detection method based on monocular images obtains depth ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V20/647G06F18/22G06F18/214
Inventor 叶晓青谭啸孙昊章宏武
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD