TW202205127A - Target detection method, electronic equipment and computer-readable storage medium - Google Patents

Abstract

The present disclosure relates to a target detection method, an electronic equipment and a computer-readable storage medium. The method includes: constructing a detection network of the target category; using the detection network of the target category to detect images to be detected to obtain the target detection result of the image to be detected; wherein the parameters of the detection network of the target category are obtained by inputting the training image of the target category into the parameter generation network. The embodiments of the present disclosure facilitate the dynamic addition of new categories.

Description

Object detection method, electronic device and computer-readable storage medium

The present invention relates to the field of computer technology, and in particular, to a target detection method, an electronic device and a computer-readable storage medium.

In related technologies, target detection relies on large-scale training data, requiring a lot of manpower and material resources to collect and label the training data, and the more categories of objects, the higher the labeling cost. In some specific scenarios, data collection is also difficult, resulting in a small number of sample images. In practical application scenarios, the requirements are often dynamic, and detection categories may need to be dynamically increased, and the number of sample images of the increased categories may be small.

The present invention provides a target detection method, an electronic device and a computer-readable storage medium.

According to an aspect of the present invention, a target detection method is provided, comprising: Build a detection network for the target category; Use the detection network of the target category to detect the image to be detected, and obtain the target detection result of the image to be detected; The parameters of the detection network of the target category are obtained by inputting the training images of the target category into the parameter generation network.

In a possible implementation, the method further includes: Obtain one or more target training sets from the image set, wherein each target training set includes training images of K categories, each category includes M training images, and K is an integer greater than 0; Based on each target training set, the parameter generation network is trained.

By training the parameter generation network with fewer samples, the parameters of the detection network can be easily obtained, and then a detection network with a smaller number of samples can be easily constructed.

In a possible implementation manner, the M training images include N support images and O query images, and N and O are integers greater than 0; the parameter generation is trained based on each target training set Internet, including: For each target training set: Input each support image of the target training set into the parameter generation network to be trained, obtain the parameters of the general detection network of the target training set, and construct the general detection network of the target training set according to the parameters of the general detection network network; Input each query image of the target training set into a feature extraction network to be trained to obtain a feature map of each query image of the target training set; Inputting the feature maps of each query image into the general detection network, respectively, to obtain a predicted label distribution result of each query image; Determine the detection loss of the general detection network according to the predicted label distribution result and the true value label of each query image; The parameter generation network to be trained is trained according to the detection loss of the general detection network.

Through the detection loss of the general detection network, convergence can be achieved quickly, so that the training of the parameter generation network can be quickly completed.

In a possible implementation manner, the inputting each support image of the target training set into a parameter generation network to be trained, to obtain the parameters of the general detection network of the target training set, including: Input each support image of the target training set into the parameter generation network to be trained respectively, and obtain the parameters of the detection network corresponding to each support image; According to the parameters of the detection network corresponding to each support image and the real category of each support image, determine the parameters of the detection network of each category of the target training set; According to the parameters of each type of detection network of the target training set, the parameters of the general detection network of the target training set are determined.

By acquiring the parameters of the general detection network of the target training set based on the multiple support images, the accuracy of the parameters of the general detection network can be improved.

In a possible implementation, the method further includes: The feature extraction network to be trained is trained according to the detection loss of the general detection network.

By training the feature extraction network with the detection loss of the general detection network, the feature discrimination ability of the feature extraction network can be improved.

In a possible implementation manner, the training of the feature extraction network to be trained according to the detection loss of the general detection network includes: Obtain the reference detection network of the target training set; inputting the feature maps of the query images into the reference detection network respectively, to obtain the reference label distribution results of the query images; Determine the detection loss of the reference detection network according to the reference label distribution result and the true value label of each query image; The feature extraction network to be trained is trained according to the detection loss of the general detection network and the detection loss of the reference detection network.

By training the feature extraction network with the detection loss of the general detection network and the detection loss of the reference detection network, the guidance of more samples to fewer samples can be realized, and the feature discrimination ability of the feature extraction network can be further improved.

In a possible implementation manner, the parameters of the reference detection network of the target training set are obtained, including: Get a randomly initialized detection network; training the randomly initialized detection network based on all query images in the target training set; The parameters of the trained detection network are determined as the reference detection network of the target training set.

By obtaining the reference detection network, the general detection network can be guided, so that the general detection network obtained by training with few samples is closer to the reference detection network obtained by training with many samples, and the loss caused by the few samples can be reduced.

In a possible implementation manner, the training of the parameter generation network to be trained according to the detection loss of the general detection network includes: According to the parameters of the general detection network of the target training set and the parameters of the reference detection network of the target training set, determine the gap loss of the general detection network; According to the detection loss and gap loss of the general detection network, the parameters to be trained are trained to generate the parameters of the network.

In this way, jointly training the parameters of the parameter generation network to be trained according to the detection loss and gap loss of the general detection network can make the detection network obtained based on the parameter generation network more accurate.

In a possible implementation, the method further includes: determining an orthogonalization loss for the generic detection network; The parameter generation network to be trained is trained according to the orthogonalization loss of the general detection network.

By making the detection networks of different classes orthogonal to each other, the discriminative ability of the model can be improved.

In a possible implementation manner, the construction of the detection network of the target category includes: obtaining training images of the target category; Input each training image of the target category into the parameter generation network respectively, to obtain the parameters of the detection network corresponding to each training sample of the target category; Determine the parameters of the detection network of the target category according to the parameters of the detection network corresponding to each training sample of the target category; According to the parameters of the detection network of the target category, the detection network of the target category is constructed.

According to an aspect of the present invention, a target detection device is provided, comprising: Build a module, configured to build a detection network for the target category; a detection module, configured to use the detection network of the target category to detect the to-be-detected image to obtain a target detection result of the to-be-detected image; Wherein, the parameters of the detection network of the target category are obtained from the input parameter generation network based on the training image of the target category.

In a possible implementation, the apparatus further includes: The acquisition module is configured to acquire one or more target training sets from the image set, wherein each target training set includes training images of K categories, each category includes M training images, and K is greater than 0 integer; The first training module is configured to train the parameter generation network based on each target training set.

In a possible implementation manner, the M training images include N support images and O query images, where N and O are integers greater than 0; the first training module is further configured as: For each target training set: Input each support image of the target training set into the parameter generation network to be trained, obtain the parameters of the general detection network of the target training set, and construct the general detection network of the target training set according to the parameters of the general detection network network; Input each query image of the target training set into a feature extraction network to be trained to obtain a feature map of each query image of the target training set; Inputting the feature maps of each query image into the general detection network, respectively, to obtain a predicted label distribution result of each query image; Determine the detection loss of the general detection network according to the predicted label distribution result and the true value label of each query image; The parameter generation network to be trained is trained according to the detection loss of the general detection network.

In a possible implementation manner, the first training module is further configured as: Input each support image of the target training set into the parameter generation network to be trained respectively, and obtain the parameters of the detection network corresponding to each support image; According to the parameters of the detection network corresponding to each support image and the real category of each support image, determine the parameters of the detection network of each category of the target training set; According to the parameters of each type of detection network of the target training set, the parameters of the general detection network of the target training set are determined.

In a possible implementation, the apparatus further includes: The second training module is configured to train the feature extraction network to be trained according to the detection loss of the general detection network.

In a possible implementation manner, the second training module is further configured as: Obtain the reference detection network of the target training set; inputting the feature maps of the query images into the reference detection network respectively, to obtain the reference label distribution results of the query images; Determine the detection loss of the reference detection network according to the reference label distribution result and the true value label of each query image; The feature extraction network to be trained is trained according to the detection loss of the general detection network and the detection loss of the reference detection network.

In a possible implementation manner, the parameters of the reference detection network of the target training set are obtained, including: Get a randomly initialized detection network; training the randomly initialized detection network based on all query images in the target training set; The parameters of the trained detection network are determined as the reference detection network of the target training set.

In a possible implementation manner, the first training module is further configured as: According to the parameters of the general detection network of the target training set and the parameters of the reference detection network of the target training set, determine the gap loss of the general detection network; According to the detection loss and gap loss of the general detection network, the parameters to be trained are trained to generate the parameters of the network.

In a possible implementation, the apparatus further includes: a determination module configured to determine the orthogonalization loss of the universal detection network; The third training module is configured to train the parameter generation network to be trained according to the orthogonalization loss of the general detection network.

In a possible implementation, the building module is further configured to: obtaining training images of the target category; Input each training image of the target category into the parameter generation network respectively, to obtain the parameters of the detection network corresponding to each training sample of the target category; Determine the parameters of the detection network of the target category according to the parameters of the detection network corresponding to each training sample of the target category; According to the parameters of the detection network of the target category, the detection network of the target category is constructed.

According to an aspect of the present invention, an electronic device is provided, comprising: a processor; a memory configured to store instructions executable by the processor; wherein the processor is configured to invoke the instructions stored in the memory to execute the above method.

According to an aspect of the present invention, there is provided a computer-readable storage medium having computer program instructions stored thereon, the computer program instructions implementing the above method when executed by a processor.

In the embodiment of the present invention, for the target category with a small number of training images, the parameters of the detection network of the target category can be obtained through the parameter generation network first, and then the detection network of the target category can be constructed according to the parameters, so as to achieve the target Class object detection. In this way, the labeling cost of training images is reduced, and the risk of overfitting caused by directly training the detection network with a small number of training images is reduced. Further, the embodiments of the present invention are conducive to dynamically adding new categories.

It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. Other features and aspects of the present invention will become apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings.

Various exemplary embodiments, features and aspects of the present invention will be described in detail below with reference to the accompanying drawings. The same reference numbers in the figures denote elements that have the same or similar functions. While various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

The term "and/or" in this article is only an association relationship to describe associated objects, indicating that there can be three kinds of relationships, for example, A and/or B, which can mean that A exists alone, A and B exist at the same time, and B exists alone. three conditions. In addition, the term "at least one" herein refers to any combination of any one of a plurality or at least two of a plurality, for example, including at least one of A, B, and C, and may mean including those composed of A, B, and C. Any one or more elements selected in the collection.

In addition, in order to better illustrate the present invention, numerous specific details are given in the following detailed description. It will be understood by those skilled in the art that the present invention may be practiced without certain specific details. In some instances, methods, means, components and circuits well known to those skilled in the art have not been described in detail so as not to obscure the subject matter of the present invention.

Object detection is a classic problem in computer vision. It mainly determines whether an image contains a certain type of object, and if it does, the position of each object needs to be given. Object detection is the cornerstone of image content understanding and the basis for many more complex visual understanding tasks, such as tracking recognition, instance segmentation, scene classification, and event detection. With the development of technology, object detection has been widely used in real life, such as face recognition, automatic driving, security control and entertainment interaction. Typically, for a detection network to learn a new class, a large number of images of that class are required. However, in practical applications, a large number of images of new categories may not be obtained. For example, after a photographer captures a rare bird (or a rare scene, a deep-sea animal, etc.), researchers may need to detect whether such a bird has appeared from a large number of bird images during the research process. At this time, it is very time-consuming and labor-intensive to manually confirm whether such small birds have appeared in the images of a large number of birds. Therefore, it is necessary to detect the target of this kind of bird through a detection network. Due to the small number of such birds and the few images confirmed to contain such birds, it is not possible to directly train a detection network that can accurately detect objects of such birds through images that have been confirmed to contain such birds. . The target detection method provided by the embodiment of the present invention can construct a relatively accurate detection network for detecting the target of the small bird based on a small number of images containing the small bird. In this way, object detection can be performed on images of a large number of birds to determine whether such small birds have appeared.

FIG. 1 shows a flowchart of a target detection method according to an embodiment of the present invention. As shown in Figure 1, the target detection method may include: Step S11, constructing a detection network of the target category. Step S12, using the detection network of the target category to detect the image to be detected, to obtain a target detection result of the image to be detected.

The parameters of the detection network of the target category are obtained by inputting the training images of the target category into the parameter generation network.

In the embodiment of the present invention, for the target category with a small number of training images, the parameters of the detection network of the target category can be obtained through the parameter generation network first, and then the detection network of the target category can be constructed according to the parameters, so as to achieve the target Class object detection. In this way, the labeling cost of training images is reduced, and the risk of overfitting caused by directly training the detection network with a small number of training images is reduced. Further, the embodiments of the present invention are conducive to dynamically adding new categories.

It can be understood that the target category can also be a category with a large number of training images, and the target detection method provided by the embodiment of the present invention can also be applied to a category with a large number of training images.

In a possible implementation manner, the target detection method may be executed by an electronic device such as a terminal device or a server, and the terminal device may be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone , wireless phones, personal digital assistants (Personal Digital Assistant, PDA), handheld devices, computing devices, vehicle-mounted devices, wearable devices, etc., the method can be implemented by the processor calling the computer-readable instructions stored in the memory . Alternatively, the method may be performed by a server.

In step S11, the target category may represent a category to be subjected to target detection. In one example, the target class may be a class with a small number of training images, eg, the target class may be a class with one or several training images. In practical application scenarios, the target category can be a dynamically added new category.

A detection network may represent a network used for object detection. In one example, the structure of the detection network may be a network capable of performing anchor-free object detection, such as an FCOS (Full Convolutional One Stage Object Detection, fully convolutional one-stage target detection) network. The embodiment of the present invention does not limit the detection network.

A detection network for object classes may represent a network for object detection for object classes. That is to say, the detection network of the target category can detect whether there is an object of the target category in the image to be detected. In this embodiment of the present invention, the parameters of the detection network of the target category may be obtained first, and then the detection network of the target category may be constructed based on the parameters of the detection network of the target category. Among them, the parameters of the detection network of the target category are obtained from the input parameter generation network based on the training image of the target category.

The parameter generation network can be used to generate the parameters of the detection network. The parameter generation network takes the training image as input, the parameter of the detection network as the output, and the training image of the target category is input into the parameter generation network, and the parameters of the detection network of the target category can be obtained. This embodiment of the present invention does not limit the structure of the parameter generation network.

In a possible implementation manner, the process of constructing the detection network is as follows: first, obtain the training images of the target category, input each training image of the target category into the parameter generation network, and obtain the the parameters of the detection network corresponding to each training sample of the target category; then, according to the parameters of the detection network corresponding to each training sample of the target category, determine the parameters of the detection network of the target category; finally, According to the parameters of the detection network of the target category, the detection network of the target category is constructed.

In the embodiment of the present invention, each training image of the target category can be input into the parameter generation network respectively, and the parameters of the detection network corresponding to each training image of the target category can be obtained. Since these training images all belong to the target category, the parameters of the detection network of the target category can be determined according to the parameters of the detection network corresponding to these training images. In one example, the parameters of the detection network corresponding to the training images of the target category may be averaged, and the averaged parameters of the detection network may be determined as the parameters of the detection network of the target category. In yet another example, first, the weight information of each training image of the target category may be determined according to information such as the position or size of the destination area (the area where the object of the target category is located) in the training image; then, based on the weight information, The parameters of the detection network corresponding to each training image of the target category are weighted and averaged, and the parameters of the detection network after the weighted average are determined as the parameters of the detection network of the target category.

After acquiring the parameters of the detection network, a corresponding detection network can be constructed based on the structure of the detection network. That is, after acquiring the parameters of the detection network of the target category, the detection network of the target category can be constructed based on the structure of the detection network.

In a possible implementation manner, the parameters may be set as the detection network of the detection network of the target category, and the detection network of the target category may be directly determined as the detection network of the target category. In this way, after the training images of the target category are input into the parameters to generate the network, the detection network of the target category can be obtained conveniently and quickly.

In a possible implementation manner, the parameters of the detection network of the target category may be set as the parameters of the detection network first, and the detection network may be determined as the initialization detection network of the target category; then, the initialization detection network may be fine-tuned to obtain Detection network for object classes. In one example, the initial detection network can be fine-tuned by loss minimization. The loss here can include detection loss and quadrature loss for initializing the detection network. Among them, the detection loss of the initialized detection network can be determined according to the predicted label distribution result and the corresponding ground-truth label output after the input of the training image of the target category to the initialized detection network.

In this way, an optimized detection network can be obtained in a shorter time, thereby improving the accuracy of the detection network of the target category.

In step S12, the image to be detected may be input into the detection network of the target category to obtain the target detection result of the image to be detected. In one example, the target detection result may include the probability that the image to be detected is of the target category and the position information of the object of the target category in the image to be detected.

In the embodiment of the present invention, a network is first generated based on parameters to obtain the parameters of the detection network of the target category, and then the detection network of the target category is constructed according to the parameters of the detection network of the target category, thereby realizing the detection network of the target category. target detection. The parameter generation network is an important tool of the target detection method of the embodiment of the present invention. The training process of the parameter generation network is described below.

In a possible implementation manner, the training process of the parameter generation network may include: acquiring one or more target training sets from an image set; and training the parameter generation network based on each target training set.

The image set includes training images of C categories (called C base categories), and each category includes A training images as an example for illustration. The process of obtaining a target training set from the image set may include: randomly selecting K categories from C categories, and randomly selecting M training images from A training images of each category. At this time, the target training set includes training images of K categories, and each category includes M training images. By repeating this process, multiple target training sets can be obtained from the image set.

Wherein, C, A, K, and M are integers greater than 0, and C>K, A>M.

The number of K and M can be set as required. Since the target detection method of the embodiment of the present invention needs to solve the problem of detecting categories with a small number of training images, the embodiment of the present invention adopts a small number of categories when training the parameter generation network, and the number of categories for each category is small. The number of training images is also smaller. In one example, K may be 5; M may be 11, 15, or 20, etc. The corresponding C can be 1000 or 2000, etc., and A can be 5000 or 10000. It can be understood that the process of generating the network for training parameters provided by the embodiment of the present invention is also applicable to a large number of categories, therefore, the number of training images of each category can be large, and M can also be 500 or 1000, etc. .

It should be noted that, for each category of the target training set, the M training images included in the category may include N support images and O query images, where N and O are integers greater than 0, and M≥N +O. In the case of M=N+O, for each category of the target training set, N training images can be randomly selected from the M training images of this category as support images, and the rest of the category can be selected as support images. training images as query images. In the case of M>N+O, for each category of the target training set, N training images can be randomly selected from the M training images of the category as support images, and the remaining training images of the category can be selected from the M training images of the category. O training images are randomly selected as query images.

The following takes a target training set as an example to illustrate the training process of the parameter generation network. The process of using multiple target training sets to train parameters to generate a network is actually a process of repeatedly using one target training set to train parameters to generate a network, which will not be repeated here.

In a possible implementation manner, training a parameter generation network based on a target training set may include: first, input each support image of the target training set into the parameter generation network to be trained, and obtain the target training set The parameters of the general detection network, and according to the parameters of the general detection network, the general detection network of the target training set is constructed; secondly, each query image of the target training set is input into the feature extraction network to be trained, and the result is obtained The feature map of each query image in the target training set; thirdly, input the feature map of each query image into the general detection network to obtain the predicted label distribution result of each query image; finally, according to The predicted label distribution result and true value label of each query image determine the detection loss of the general detection network, and train the parameter generation network to be trained according to the detection loss of the general detection network.

Wherein, inputting each support image of the target training set into the parameter generation network to be trained, and obtaining the parameters of the general detection network of the target training set, may include: inputting each support image of the target training set to the to-be-trained The trained parameter generation network obtains the parameters of the detection network corresponding to each support image; according to the parameters of the detection network corresponding to each support image and the real category of each support image, each target training set is determined. The parameters of the detection network of the target training set are determined; and the parameters of the general detection network of the target training set are determined according to the parameters of the detection network of each type of the target training set.

In one example, the parameters of the detection network corresponding to the support images of the same category may be averaged or weighted averaged according to the real category of each support image (the weight may be based on the location or size of the destination area in the support image) and other information to be determined) to obtain the parameters of the detection network of the corresponding category. Then, the parameters of each category of detection network are spliced into the parameters of the general detection network of the target training set.

201和特徵提取網路202

。其中，參數生成網路

的參數為

，特徵提取網路

的參數為

。FIG. 2 shows a schematic diagram of a network architecture according to an embodiment of the present invention. As shown in FIG. 2, the network architecture 200 includes a parameter generation network

201 and Feature Extraction Networks 202

. Among them, the parameter generation network

The parameters are

, the feature extraction network

The parameters are

.

，該目標訓練集包括支援集

和查詢集

。As shown in Figure 2, the target training set was obtained from the image set

, the target training set includes the support set

and queryset

.

包括K個類別的支援圖像，每個類別包括N個支援圖像。

表示支援圖像中的目的地區域，

表示

的真值標籤，

表示支持集

中第i個支援圖像的目的地區域和真值標籤，

，

，其中，

表示

的類別，

表示

的位置資訊。Among them, the support set

K categories of support images are included, and each category includes N support images.

Indicates the destination area in the support image,

Express

The truth label of ,

Indicates the support set

the destination region and ground truth label of the i-th support image,

,

,in,

Express

category,

Express

location information.

包括K個類別的查詢圖像，每個類別包括O個查詢圖像。

表示查詢圖像中的目的地區域，

表示

的真值標籤，

表示查詢集

中第j個查詢圖像的目的地區域和真值標籤，

。

，其中，

表示

的類別，

表示

的位置資訊。query set

Query images of K categories are included, and each category includes O query images.

represents the destination area in the query image,

Express

The truth label of ,

Represents a queryset

The destination region and ground-truth label of the j-th query image in ,

.

,in,

Express

category,

Express

location information.

With reference to Fig. 2, using the target training set D, the process of training the parameter generation network may include the following.

中的各支援圖像進行裁剪得到各支援圖像的目的地區域

（在一個示例中，目的地區域的尺寸可以為224圖元*224圖元），將各支援圖像的目的地區域

輸入待訓練的參數生成網路

中，可以得到每個支援圖像對應的檢測網路的參數，將同一類別的支援圖像

對應的檢測網路的參數進行平均（或者加權平均），可以得到該類別的檢測網路的參數。公式（1）示出了類別k的檢測網路的參數：

（1）； 其中，D表示檢測網路的參數的維度，

表示類別k的檢測網路的參數；

，也就是說檢測網路的類別與支援圖像的類別一致。(1) Build a general detection network for the target training set, specifically: the support set

Crop each support image in to get the destination area of each support image

(In an example, the size of the destination area can be 224 primitives*224 primitives), the destination area of each supporting image

Enter the parameters to be trained to generate the network

, the parameters of the detection network corresponding to each support image can be obtained, and the support images of the same category can be

The parameters of the corresponding detection network are averaged (or weighted average) to obtain the parameters of the detection network of this category. Equation (1) shows the parameters of the detection network for category k:

(1); Among them, D represents the dimension of the parameters of the detection network,

Represents the parameters of the detection network of category k;

, that is to say, the category of the detection network is consistent with the category of the supported image.

，然後根據該通用檢測網路的參數

，可以構建出目標訓練集的通用檢測網路。By splicing the parameters of the K categories of detection networks, the parameters of the general detection network can be obtained.

, and then based on the general detection network parameters

, a general detection network for the target training set can be constructed.

中的各查詢圖像進行裁剪後得到各查詢圖像的目的地區域

（在一個示例中，目的地區域的短邊為600圖元，長邊不超過1000圖元）輸入特徵提取網路

中，得到各查詢圖像的特徵圖

。(2) Obtain the feature map of the query image, specifically: the query set

The destination area of each query image is obtained by cropping each query image in

(In one example, the short side of the destination area is 600 primitives and the long side does not exceed 1000 primitives) Input Feature Extraction Network

, the feature map of each query image is obtained

.

輸入參數為

的通用檢測網路中，可以得到各查詢圖像的預測標籤分佈結果。根據各查詢圖像的預測標籤分佈結果和真值標籤

，可以得到通用檢測網路的檢測損失。在一個示例中，可以通過公式（2）得到通用檢測網路的檢測損失。

（2）； 其中，

表示通用檢測網路的檢測損失，loss（…）表示損失函數。

表示以查詢圖像的真值標籤

和查詢圖像的特徵圖

輸入參數為

的通用檢測網路中得到的預測標籤分佈結果為參數的損失函數。本發明實施例中對損失函數的結構不做限制，例如可以為均方誤差函數、交叉熵函數等。(3) Determine the detection loss of the general detection network, specifically: the feature map of each query image

The input parameters are

In the general detection network of , the predicted label distribution results of each query image can be obtained. According to the predicted label distribution results and ground-truth labels of each query image

, the detection loss of the general detection network can be obtained. In one example, the detection loss of the general detection network can be obtained by Eq. (2).

(2); wherein,

represents the detection loss of the generic detection network, and loss(…) represents the loss function.

Representation to query the ground truth label of an image

and the feature map of the query image

The input parameters are

The predicted label distribution obtained in the general detection network results in a loss function of the parameters. The structure of the loss function is not limited in the embodiment of the present invention, for example, it may be a mean square error function, a cross entropy function, or the like.

，訓練待訓練的參數生成網路

。(4) According to the detection loss of the general detection network

, train the parameter generation network to be trained

.

的參數

，以實現對參數生成網路

的訓練。To minimize the detection loss of the general detection network, adjust the parameters to generate the network

the parameters

, to generate a network for the parameters

training.

，可以用來生成新類別的檢測網路的參數，具有將其檢測網路生成能力轉移到新類上的潛力。In this way, the network is generated by parameters trained with a small number of samples.

, parameters that can be used to generate detection networks for new classes, with the potential to transfer their detection network generation capabilities to new classes.

In a possible implementation manner, the method further includes: training the feature extraction network to be trained according to the detection loss of the general detection network.

的過程中，可以同時對特徵提取網路

進行訓練。也就是說，還可以以通用檢測網路的檢測損失最小化為目標，更新特徵提取網路

的參數

。From the network architecture shown in Figure 2, it can be seen that the network is generated in the training parameters.

During the process, the feature extraction network can be simultaneously

to train. That is to say, it is also possible to update the feature extraction network with the goal of minimizing the detection loss of the general detection network.

the parameters

.

In a possible implementation manner, according to the detection loss of the general detection network, training the feature extraction network to be trained includes: acquiring parameters of the reference detection network of the target training set; training according to the target The parameters of the reference detection network of the target training set are constructed, and the reference detection network of the target training set is constructed; the feature maps of the query images are input into the reference detection network respectively, and the reference labels of the query images are obtained. distribution result; according to the reference label distribution result and true value label of each query image, determine the detection loss of the reference detection network; according to the detection loss of the general detection network and the detection of the reference detection network loss, training the feature extraction network to be trained.

The reference detection network may be used to represent the detection network obtained by training based on training images of all categories of the image set.

的區分能力被限制在各目標訓練集涉及到的類別中，使其提取特徵的能力減弱。同時，採用目標訓練集訓練參考檢測網路和特徵提取網路時，訓練過程涉及的訓練圖像的數量較少。而採用少量訓練圖像訓練得到的通用檢測網路相較於採用大量訓練圖像得到的檢測網路的目標檢測能力較弱。因此，在本發明實施例中，引入通過多種類別的大量訓練圖像訓練得到的參考檢測網路，對參數生成網路

和特徵提取網路

的訓練進行優化。When using the target training set to train the reference detection network and the feature extraction network, only K categories are involved in one training process, and multiple training is still limited to a limited number of categories. In this way, the trained feature extraction network will

The discriminative ability is limited to the categories involved in each target training set, which weakens its ability to extract features. At the same time, when using the target training set to train the reference detection network and the feature extraction network, the number of training images involved in the training process is small. The general detection network trained with a small number of training images has weaker target detection ability than the detection network obtained with a large number of training images. Therefore, in the embodiment of the present invention, a reference detection network obtained by training a large number of training images of various categories is introduced, and the parameter generation network is

and feature extraction network

training is optimized.

輸入參數為θ的參考檢測網路中，可以得到各查詢圖像的參考標籤分佈結果。根據各查詢圖像的參考標籤分佈結果和真值標籤

，可以得到參考檢測網路的檢測損失。在一個示例中，可以通過公式（3）示得到參考檢測網路的檢測損失。

（3）； 其中，

表示參考檢測網路的檢測損失，loss（…）表示損失函數。

表示以查詢圖像的真值標籤

和查詢圖像的特徵圖

輸入參數為θ的參考檢測網路中得到的參考標籤分佈結果為參數的損失函數。本發明實施例中對損失函數的結構不做限制，例如可以為均方誤差函數、交叉熵函數等。FIG. 3 shows a schematic diagram of a network architecture according to an embodiment of the present invention. The network architecture shown in FIG. 3 adds a reference detection network 301 whose parameter is θ on the basis of FIG. 2 . The feature map of each query image

In the reference detection network whose input parameter is θ, the reference label distribution results of each query image can be obtained. Distribution results and ground-truth labels based on the reference labels of each query image

, the detection loss of the reference detection network can be obtained. In one example, the detection loss of the reference detection network can be obtained by formula (3).

(3); wherein,

represents the detection loss of the reference detection network, and loss(…) represents the loss function.

Representation to query the ground truth label of an image

and the feature map of the query image

The reference label distribution result obtained in the reference detection network with input parameter θ is the loss function of the parameter. The structure of the loss function is not limited in the embodiment of the present invention, for example, it may be a mean square error function, a cross entropy function, or the like.

和公式（3）中的

均可以查詢圖像的真值標籤，區別是

是K個類別中的一個，

是所有類別中的一個。It should be noted that in formula (2)

and in formula (3)

Both can query the ground truth label of the image, the difference is

is one of K categories,

is one of all categories.

In this way, since the reference detection network is trained based on training images of all categories, the feature extraction to be trained is jointly trained according to the detection loss of the general detection network and the detection loss of the reference detection network. The network can improve the feature discrimination ability of the feature extraction network.

In a possible implementation manner, obtaining the parameters of the reference detection network of the target training set may include: obtaining a randomly initialized detection network; Carry out training; determine the parameters of the detection network after training as the reference detection network of the target training set.

和特徵提取網路

可以同時進行訓練。基於該目標訓練集的所有查詢圖像對所述隨機初始化的檢測網路進行訓練的過程可以參照相關技術中訓練檢測網路的訓練方法，例如YOLO、SSD等，對此本發明不做限制。First, a detection network is randomly initialized as the detection network to be trained, and then based on all the query images of the target training set, the detection network to be trained is obtained, and the reference detection network of the target training set is obtained. Reference detection network and parameter generation network for the target training set

and feature extraction network

Can be trained at the same time. For the process of training the randomly initialized detection network based on all the query images of the target training set, reference may be made to the training methods for training the detection network in the related art, such as YOLO, SSD, etc., which is not limited in the present invention.

，其中，D為參考檢測網路的參數的維度。將K個類別的參考檢測網路的參數進行拼接，可以得到目標訓練集的參考檢測網路的參數

。The parameters of the reference detection network of the target training set are also obtained by splicing the parameters of the detection networks of the K categories. Based on the parameters of the reference detection network of the target training set, the reference detection network of the target training set can be constructed. For example, suppose the parameters of the reference detection network for category k in the K categories of the target training set are

, where D is the dimension of the parameters of the reference detection network. By splicing the parameters of the K categories of reference detection networks, the parameters of the reference detection network of the target training set can be obtained

.

It should be noted that, in the embodiment of the present invention, a data set including K categories may also be reconstructed for training of the reference detection network. For the training process, reference may be made to the above-mentioned training process using query images, which will not be repeated here.

In a possible implementation manner, according to the detection loss of the general detection network, training the parameter generation network to be trained includes: according to the parameters of the general detection network of the target training set and the parameters of the target training set Referring to the parameters of the detection network, the gap loss of the general detection network is determined; according to the detection loss and gap loss of the general detection network, the parameters to be trained are trained to generate the parameters of the network.

（4）；

（5）； 其中，

和

為通用檢測網路的差距損失

的兩個表現形式。

是一個條件函數，括弧中條件為真時，取值為1，括弧中條件為假時，取值為0。

表示一階範數，

表示二階範數。

和

分別表示類別c對應的通用檢測網路和參考檢測網路的參數。In one example, the gap loss of the general detection network can be obtained by Equation (4) or Equation (5).

(4);

(5); wherein,

and

Gap Loss for Universal Detection Networks

two manifestations.

is a conditional function that takes the value 1 when the condition in parentheses is true, and takes the value 0 when the condition in parentheses is false.

represents the first-order norm,

represents the second-order norm.

and

Represent the parameters of the general detection network and the reference detection network corresponding to category c, respectively.

In this way, since the reference detection network is trained based on training images of all categories, according to the detection loss and gap loss of the general detection network, jointly training the parameters of the to-be-trained parameter generation network can be The accuracy of the detection network obtained based on the parameter generation network is higher.

In a possible implementation manner, the method may further include: determining an orthogonalization loss of the general detection network; training the parameter generation network to be trained according to the orthogonalization loss of the general detection network road.

（6）； 其中，

是

的行標準化版本，

表示1階範數，I是單位矩陣。In one example, the orthogonalization loss of the generic detection network can be determined by Equation (6).

(6); wherein,

Yes

The row normalized version of ,

represents the norm of order 1, and I is the identity matrix.

In the embodiment of the present invention, by making detection networks of different types orthogonal to each other, the distinguishing ability of the model can be improved.

、特徵提取網路

和參考檢測網路可以同時進行訓練。因此，本發明實施例中可以通過公式（7），確定一個總的訓練損失。

（7）； 其中，L表示總的訓練損失，

表示通用檢測網路的檢測損失（參見公式（2）），

表示參考檢測網路的檢測損失（參見公式（3）），

表示通用檢測網路的差距損失（參見公式（4）和公式（5）），

表示通用檢測網路的正交損失（參見公式（6））。

和

為超參數。

和

可以根據需要進行設置。在一個示例中，

可以取0.01，

可以取1。Considering that in the embodiment of the present invention, the parameter generation network shown in FIG. 3

, feature extraction network

and the reference detection network can be trained simultaneously. Therefore, in this embodiment of the present invention, a total training loss can be determined by formula (7).

(7); where L represents the total training loss,

represents the detection loss of the general detection network (see Equation (2)),

represents the detection loss of the reference detection network (see Equation (3)),

represents the gap loss of the generic detection network (see Equation (4) and Equation (5)),

represents the quadrature loss of the generic detection network (see Equation (6)).

and

are hyperparameters.

and

Can be set as required. In one example,

You can take 0.01,

1 can be taken.

、特徵提取網路

和參考檢測網路進行訓練，調整參數

、

和θ。In this embodiment of the present invention, a network can be generated for parameters based on L at the same time

, feature extraction network

Train with the reference detection network and adjust the parameters

,

and θ.

It can be understood that the above method embodiments mentioned in the present invention can be combined with each other to form a combined embodiment without violating the principle and logic. Due to space limitations, the present invention will not repeat them. Those skilled in the art can understand that, in the above method of the specific embodiment, the specific execution order of each step should be determined by its function and possible internal logic.

In addition, the present invention also provides a target detection device, an electronic device, a computer-readable storage medium, and a program, all of which can be used to implement any target detection method provided by the present invention. For the corresponding technical solutions and descriptions, refer to the corresponding records in the Methods section, No longer.

FIG. 4 shows a block diagram of a target detection apparatus according to an embodiment of the present invention. As shown in Figure 4, the device 40 includes: The construction module 41 is configured to construct the detection network of the target category; The detection module 42 is configured to use the detection network of the target category to detect the to-be-detected image to obtain the target detection result of the to-be-detected image; Wherein, the parameters of the detection network of the target category are obtained from the input parameter generation network based on the training image of the target category.

In a possible implementation, the apparatus further includes: The acquisition module is configured to acquire one or more target training sets from the image set, wherein each target training set includes training images of K categories, each category includes M training images, and K is greater than 0 integer; The first training module is configured to train the parameter generation network based on each target training set.

In a possible implementation manner, the M training images include N support images and O query images, where N and O are integers greater than 0; the first training module is further configured as: For each target training set: Input each support image of the target training set into the parameter generation network to be trained, obtain the parameters of the general detection network of the target training set, and construct the general detection network of the target training set according to the parameters of the general detection network network; Input each query image of the target training set into a feature extraction network to be trained to obtain a feature map of each query image of the target training set; Inputting the feature maps of each query image into the general detection network, respectively, to obtain a predicted label distribution result of each query image; Determine the detection loss of the general detection network according to the predicted label distribution result and the true value label of each query image; The parameter generation network to be trained is trained according to the detection loss of the general detection network.

In a possible implementation manner, each support image of the target training set is input into the parameter generation network to be trained, and the parameters of the general detection network of the target training set are obtained, including: Input each support image of the target training set into the parameter generation network to be trained respectively, and obtain the parameters of the detection network corresponding to each support image; According to the parameters of the detection network corresponding to each support image and the real category of each support image, determine the parameters of the detection network of each category of the target training set; According to the parameters of each type of detection network of the target training set, the parameters of the general detection network of the target training set are determined.

In a possible implementation, the apparatus further includes: The second training module is configured to train the feature extraction network to be trained according to the detection loss of the general detection network.

In a possible implementation manner, the second training module is further configured as: Obtain the reference detection network of the target training set; inputting the feature maps of the query images into the reference detection network respectively, to obtain the reference label distribution results of the query images; Determine the detection loss of the reference detection network according to the reference label distribution result and the true value label of each query image; The feature extraction network to be trained is trained according to the detection loss of the general detection network and the detection loss of the reference detection network.

In a possible implementation manner, the parameters of the reference detection network of the target training set are obtained, including: Get a randomly initialized detection network; training the randomly initialized detection network based on all query images in the target training set; The parameters of the trained detection network are determined as the reference detection network of the target training set.

In a possible implementation manner, according to the detection loss of the general detection network, training the parameter generation network to be trained includes: According to the parameters of the general detection network of the target training set and the parameters of the reference detection network of the target training set, determine the gap loss of the general detection network; According to the detection loss and gap loss of the general detection network, the parameters to be trained are trained to generate the parameters of the network.

In a possible implementation, the apparatus further includes: a determination module configured to determine the orthogonalization loss of the universal detection network; The third training module is configured to train the parameter generation network to be trained according to the orthogonalization loss of the general detection network.

In a possible implementation, the building module is further configured to: obtaining training images of the target category; Input each training image of the target category into the parameter generation network respectively, to obtain the parameters of the detection network corresponding to each training sample of the target category; Determine the parameters of the detection network of the target category according to the parameters of the detection network corresponding to each training sample of the target category; According to the parameters of the detection network of the target category, the detection network of the target category is constructed.

In some embodiments, the functions or modules included in the apparatus provided in the embodiments of the present invention may be used to execute the methods described in the above method embodiments. For specific implementation, reference may be made to the above method embodiments. For brevity, I won't go into details here.

An embodiment of the present invention further provides a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above-mentioned method is implemented. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

An embodiment of the present invention further provides an electronic device, including: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute the above method.

Embodiments of the present invention also provide a computer program product, including computer-readable codes. When the computer-readable codes are run on a device, a processor in the device executes instructions for implementing the target detection method provided in any of the above embodiments. .

Embodiments of the present invention further provide another computer program product for storing computer-readable instructions, and when the instructions are executed, the computer executes the operations of the target detection method provided by any of the above embodiments.

The electronic device may be provided as a terminal, server or other form of device.

FIG. 5 shows a block diagram of an electronic device 800 according to an embodiment of the present invention. For example, the electronic device 800 may be a mobile phone, a computer, a digital broadcasting terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, or the like.

5, an electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensing server component 814, and communication component 816.

The processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 can include one or more processors 820 to execute instructions to perform all or some of the steps of the methods described above. Additionally, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

The memory 804 is configured to store various types of data to support the operation of the electronic device 800 . Examples of such data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, and the like. Memory 804 may be implemented by any type of volatile or non-volatile storage device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Disk or CD.

Power supply assembly 806 provides power to various components of electronic device 800 . Power supply components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800 .

Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, swipe, and gestures on the touch panel. The touch sensor may not only sense the boundaries of a touch or swipe action, but also detect the duration and pressure associated with the touch or swipe action. In some embodiments, the multimedia component 808 includes a front-facing camera and/or a rear-facing camera. When the electronic device 800 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera can receive external multimedia data. Each of the front and rear cameras can be a fixed optical lens system or have focal length and optical zoom capability.

Audio component 810 is configured to output and/or input audio signals. For example, audio component 810 includes a microphone (MIC) that is configured to receive external audio signals when electronic device 800 is in operating modes, such as call mode, recording mode, and voice recognition mode. The received audio signal may be further stored in memory 804 or transmitted via communication component 816 . In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing element 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, and the like. These buttons may include, but are not limited to: home button, volume buttons, start button, and lock button.

Sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for electronic device 800 . For example, the sensor assembly 814 can detect the open/closed state of the electronic device 800, the relative positioning of the components, such as the display and keypad of the electronic device 800, the sensor assembly 814 can also detect the electronic device 800 or Changes in the position of a component of the electronic device 800 , presence or absence of user contact with the electronic device 800 , orientation or acceleration/deceleration of the electronic device 800 and changes in the temperature of the electronic device 800 . Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. The sensor assembly 814 may also include a light sensor, such as a complementary metal oxide semiconductor (CMOS) or charge coupled device (CCD) image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. The electronic device 800 may access a wireless network based on a communication standard, such as wireless network (WiFi), second generation mobile communication technology (2G) or third generation mobile communication technology (3G), or a combination thereof. In an exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology and other technologies.

In an exemplary embodiment, electronic device 800 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation for performing the above method.

In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions executable by the processor 820 of the electronic device 800 to accomplish the above method.

FIG. 6 shows a block diagram of an electronic device 1900 according to an embodiment of the present invention. For example, the electronic device 1900 may be provided as a server. 6, the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions executable by the processing component 1922, such as applications. An application program stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Additionally, the processing component 1922 is configured to execute instructions to perform the above-described methods.

The electronic device 1900 may also include a power supply assembly 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input output (I/O) Interface 1958. The electronic device 1900 can operate an operating system based on the memory 1932, such as Microsoft Server Operating System (Windows Server ^TM ), a graphical user interface based operating system (Mac OS X ^TM ) introduced by Apple Inc. Process Computer Operating System (Unix ^TM ), Free and Open Source Unix-like Operating System (Linux ^TM ), Open Source Unix-like Operating System (FreeBSD ^TM ) or the like.

In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions executable by the processing component 1922 of the electronic device 1900 to accomplish the above method.

The present invention may be a system, method and/or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the present invention.

A computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device. The computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (non-exhaustive list) of computer readable storage media include: portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable Design read only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read only memory (CD-ROM), digital versatile disk (DVD), memory stick , a floppy disk, a mechanically encoded device, such as a punched card or a raised structure in a groove with instructions stored thereon, and any suitable combination of the foregoing. As used herein, computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (eg, light pulses through fiber optic cables), or Electrical signals carried by wires.

The computer-readable program instructions described herein may be downloaded from computer-readable storage media to various computing/processing devices, or downloaded to external computers or external storage over a network, such as the Internet, a local area network, a wide area network, and/or a wireless network device. Networks may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. A network interface card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for computer-readable storage stored in each computing/processing device in the medium.

The computer program instructions for carrying out the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or any other information in one or more programming languages. Combining source or object code written in programming languages including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as the "C" language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely remotely. run on a client computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any kind of network—including a local area network (LAN) or a wide area network (WAN)—or, can be connected to an external computer (for example, using the Internet road service provider to connect via the Internet). In some embodiments, electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), are personalized by utilizing state information of computer readable program instructions. Electronic circuits may execute computer readable program instructions to implement various aspects of the present invention.

Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.

These computer readable program instructions may be provided to the processor of a general purpose computer, special purpose computer or other programmable data processing device to produce a machine for execution of the instructions by the processor of the computer or other programmable data processing device When, means are created that implement the functions/acts specified in one or more of the blocks in the flowchart and/or block diagrams. These computer readable program instructions may also be stored on a computer readable storage medium, the instructions causing the computer, programmable data processing device and/or other equipment to operate in a particular manner, so that the computer readable medium storing the instructions Included is an article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.

Computer readable program instructions can also be loaded into a computer, other programmable data processing device, or other equipment, so that a series of operational steps are performed on the computer, other programmable data processing device, or other equipment to generate a computer Processes of implementation such that instructions executing on a computer, other programmable data processing apparatus, or other device implement the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions that contains one or more logic for implementing the specified logic Executable instructions for the function. In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by dedicated hardware-based systems that perform the specified functions or actions. implementation, or may be implemented in a combination of special purpose hardware and computer instructions.

The computer program product can be implemented by hardware, software or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK), etc. Wait.

Various embodiments of the present invention have been described above, and the foregoing descriptions are exemplary, not exhaustive, and not limiting of the disclosed embodiments. Numerous modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the various embodiments, the practical application or improvement over the technology in the marketplace, or to enable others of ordinary skill in the art to understand the various embodiments disclosed herein.

Industrial Applicability The present invention provides a target detection method, an electronic device and a computer-readable storage medium, wherein a detection network of a target category is constructed; an image to be detected is detected by using the detection network of the target category to obtain the to-be-detected image The target detection result of the image; wherein, the parameters of the detection network of the target category are obtained from the input parameter generation network based on the training image of the target category.

202:特徵提取網路

40:目標檢測裝置 41:構建模組 42:檢測模組 800:電子設備 802:處理組件 804:記憶體 806:電源組件 808:多媒體組件 810:音頻組件 812:輸入/輸出介面 814:感測器組件 816:通信組件 820:處理器 1900:電子設備 1922:處理組件 1926:電源組件 1932:記憶體 1950:網路介面 1958:輸入輸出介面 S11~S12:步驟201: Parameter generation network

202: Feature Extraction Network

40: Object Detection Device 41: Building Module 42: Detection Module 800: Electronic Equipment 802: Processing Component 804: Memory 806: Power Component 808: Multimedia Component 810: Audio Component 812: Input/Output Interface 814: Sensor Component 816: Communication Component 820: Processor 1900: Electronic Equipment 1922: Processing Component 1926: Power Component 1932: Memory 1950: Network Interface 1958: I/O Interface S11~S12: Steps

The accompanying drawings, which are incorporated into and constitute a part of this specification, illustrate embodiments consistent with the present invention, and together with the description, serve to explain the technical solutions of the present invention. 1 shows a flowchart of a target detection method according to an embodiment of the present invention; 2 shows a schematic diagram of a network architecture according to an embodiment of the present invention; 3 shows a schematic diagram of a network architecture according to an embodiment of the present invention; 4 shows a block diagram of a target detection apparatus according to an embodiment of the present invention; FIG. 5 shows a block diagram of an electronic device 800 according to an embodiment of the present invention; FIG. 6 shows a block diagram of an electronic device 1900 according to an embodiment of the present invention.

S11~S12: Steps

Claims (12)

A target detection method, comprising: Build a detection network for the target category; Use the detection network of the target category to detect the image to be detected, and obtain the target detection result of the image to be detected; The parameters of the detection network of the target category are obtained by inputting the training images of the target category into the parameter generation network. The method according to claim 1, further comprising: Obtain one or more target training sets from the image set, wherein each target training set includes training images of K categories, each category includes M training images, and K is an integer greater than 0; Based on each target training set, the parameter generation network is trained. The method according to claim 2, wherein the M training images include N support images and O query images, and N and O are integers greater than 0; Generate a network based on the above parameters, including: For each target training set: Input each support image of the target training set into the parameter generation network to be trained, obtain the parameters of the general detection network of the target training set, and construct the general detection network of the target training set based on the parameters of the general detection network network; Input each query image of the target training set into the feature extraction network to be trained to obtain a feature map of each query image of the target training set; Inputting the feature maps of the query images into the general detection network respectively, to obtain the predicted label distribution results of the query images; Determine the detection loss of the general detection network according to the predicted label distribution result and the true value label of each query image; The parameter generation network to be trained is trained according to the detection loss of the general detection network. The method according to claim 3, wherein the inputting each support image of the target training set into a parameter generation network to be trained to obtain the parameters of the general detection network of the target training set, including: Input each support image of the target training set into the parameter generation network to be trained respectively, and obtain the parameters of the detection network corresponding to each support image; According to the parameters of the detection network corresponding to each support image and the real category of each support image, determine the parameters of the detection network of each category of the target training set; According to the parameters of each type of detection network of the target training set, the parameters of the general detection network of the target training set are determined. The method according to claim 3 or 4, further comprising: The feature extraction network to be trained is trained according to the detection loss of the general detection network. The method according to claim 5, wherein the training of the feature extraction network to be trained according to the detection loss of the general detection network includes: Obtain the reference detection network of the target training set; inputting the feature maps of the query images into the reference detection network respectively, to obtain the reference label distribution results of the query images; Determine the detection loss of the reference detection network according to the reference label distribution result and the true value label of each query image; The feature extraction network to be trained is trained according to the detection loss of the general detection network and the detection loss of the reference detection network. The method according to claim 6, wherein the acquiring the parameters of the reference detection network of the target training set includes: Get a randomly initialized detection network; training the randomly initialized detection network based on all query images in the target training set; The parameters of the trained detection network are determined as the reference detection network of the target training set. The method according to claim 6, wherein the training of the parameter generation network to be trained according to the detection loss of the general detection network comprises: According to the parameters of the general detection network of the target training set and the parameters of the reference detection network of the target training set, determine the gap loss of the general detection network; According to the detection loss and gap loss of the general detection network, the parameters to be trained are trained to generate the parameters of the network. The method according to claim 3 or 4, further comprising: determining an orthogonalization loss for the generic detection network; The parameter generation network to be trained is trained according to the orthogonalization loss of the general detection network. The method according to claim 2, wherein the constructing the detection network of the target category includes: obtaining training images of the target category; Input each training image of the target category into the parameter generation network respectively, to obtain the parameters of the detection network corresponding to each training sample of the target category; Determine the parameters of the detection network of the target category according to the parameters of the detection network corresponding to each training sample of the target category; According to the parameters of the detection network of the target category, the detection network of the target category is constructed. . An electronic device comprising: processor; memory for storing processor-executable instructions; Wherein, the processor is configured to invoke the instructions stored in the memory to execute the method of any one of request items 1 to 10. A computer-readable storage medium on which computer program instructions are stored, the computer program instructions implement the method described in any one of claim 1 to 10 when the computer program instructions are executed by a processor.

Images