Instance segmentation model training method and device and instance segmentation method

A technology for segmentation models and training methods, applied in neural learning methods, biological neural network models, image analysis, etc., can solve the problems of low prediction accuracy and slow training process of instance segmentation models

Pending Publication Date: 2020-10-20
SUNING CLOUD COMPUTING CO LTD
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing instance segmentation model has a large number of network layers when extracting image features, so the whole training process is relatively slow when the amount of data is large, and the existing instance segmentation mo

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
  • Instance segmentation model training method and device and instance segmentation method
  • Instance segmentation model training method and device and instance segmentation method
  • Instance segmentation model training method and device and instance segmentation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] Such as figure 1 As shown, a training method of an instance segmentation model includes the following steps:

[0065] S11. Construct a deep learning model;

[0066] This application builds a basic network structure based on the YOLACT model, and constructs a deep learning model. The deep learning model includes: convolutional layer, activation layer, pooling layer, fully connected layer, etc. The specific structure of the model is as follows: figure 2 As shown, it includes resent18 network, FPN network, two network branches (protonet and Pred_heads) connected to FPN network, crop network, etc. Among them, the resent18 network is used to extract features, the FPN network is used to fuse features, the protonet network branch is used to segment the feature map, and the segmentation results including the foreground and background are obtained, and the Pred_heads are used to predict the feature map to obtain information about Object detection boxes, categories, confidenc...

Embodiment 2

[0111] Based on the instance segmentation model trained in the above-mentioned embodiment 1, the embodiment of the present invention also provides an instance segmentation method, such as image 3 As shown, the methods include:

[0112] S31. Obtain the picture to be detected;

[0113] S32. Input the picture to be detected to the pre-trained instance segmentation model for recognition, and output the detection frame of the picture to be detected and the result of instance segmentation.

[0114] Wherein, the identification process of the picture to be detected can refer to the training process of the model in Embodiment 1 for details. Before outputting the detection frame and instance segmentation results of the picture to be detected, it is necessary to compare the confidence with the preset value. For details, refer to figure 2 , after the Crop module compares the confidence level with the preset value, it outputs the detection frame corresponding to the confidence level hi...

Embodiment 3

[0123] Based on the above-mentioned embodiment 1, the embodiment of the present invention also provides a training device for an instance segmentation model, such as Figure 4 As shown, the device includes:

[0124] The pruning module 41 is used for pruning the pre-built deep learning model;

[0125] Acquisition module 42, is used for obtaining training set; Training set is the collection of the RGBD image with target object under a scene that different depth cameras gather, and RGBD image comprises depth map and color map;

[0126] A preprocessing module 43, configured to mark the training set;

[0127] The training module 44 is configured to use the labeled training set to train the pruned deep learning model to obtain an instance segmentation model.

[0128] Further, the preprocessing module 43 is also used to preprocess the training set before labeling, specifically including:

[0129] Perform 3D reconstruction according to the obtained depth map in the training set to ...

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

Embodiments of the invention disclose an instance segmentation model training method and device, and an instance segmentation method. The method comprises: pruning a pre-constructed deep learning model; obtaining and marking a training set, wherein the training set is a set of RGBD images with target objects in a scene collected by different depth cameras, and the RGBD images comprise depth imagesand color images; and training the pruned deep learning model by using the labeled training set to obtain an instance segmentation model. According to the invention, pruning is carried out on the network structure of the existing instance segmentation model, the whole model is lighter, the training speed of the model and the prediction speed of the model are improved, meanwhile, in order to prevent the reduction of the model prediction precision caused by the reduction of a network layer, a depth map is added, the number of channels is increased, and the training precision and the predictionprecision of the model are improved.

Description

technical field [0001] The invention belongs to the field of target detection, and in particular relates to a training method, device and instance segmentation method of an instance segmentation model. Background technique [0002] With the continuous improvement of the level of science and technology, the technology in the field of artificial intelligence has continued to mature and its application has been implemented, which has greatly improved the quality of people's lives. Nowadays, there are a large number of image and video acquisition systems in many scenarios. Applying the advanced technology in the field of artificial intelligence to the image and video system can greatly improve the system's ability to understand image and video content. , public places and other scenarios, providing intelligent monitoring technology capabilities. However, the existing instance segmentation model has a large number of network layers when extracting image features, so the whole tr...

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
IPC IPC(8): G06T7/11G06N3/04G06N3/08G06T7/194
CPCG06T7/11G06T7/194G06N3/082G06T2207/10016G06T2207/10024G06N3/045
Inventor 卢运西徐兆坤黄银君
Owner SUNING CLOUD COMPUTING CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap