Small sample point cloud semantic segmentation method and device, computer equipment and storage medium

Abstract

The application relates to a small sample point cloud semantic segmentation method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a support set point cloud and a corresponding mask query set point cloud, extracting features of the support set point cloud and the query set point cloud to obtain support set features and query set features; aligning the support set features according to the distance between the support set features and the query set features to obtain aligned support set features; obtaining corresponding prototypes according to the mask of the aligned support set features, splicing the prototypes and the query set features to obtain expanded query set features, extracting point cloud feature distribution information from the expanded query set features, correcting the point cloud feature distribution information, and segmenting corrected prototypes and corrected query set features from the corrected point cloud feature distribution information. The application can reduce the reasoning time and greatly reduce the point cloud segmentation error.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of computer vision, and particularly relates to a small sample point cloud semantic segmentation method and device, a computer device, and a storage medium. BACKGROUND

[0002] The small sample point cloud semantic segmentation method refers to a method in which, under the condition of a small amount of labeled new class samples being given, a model can perform semantic segmentation on point clouds to obtain point clouds belonging to a new class. In order to extract effective point cloud features, researchers strive to improve the accuracy of prototypes. Multiple prototypes are redundant and take a lot of time in the inference process. Due to the complex structure of point clouds, there are usually inter-class distribution deviations and intra-class distribution deviations between ideal feature distributions and actual feature distributions. Specifically, due to the distribution shift caused by the difference between support data and query data, there are inter-class distribution deviations, and due to the different local structures of classes, such as seats and legs on a chair, the characteristics of query data can be scattered, which means that the query characteristics are not closely around the relevant prototypes, resulting in intra-class distribution deviations. Due to the complex feature distribution of point clouds, how to reduce these deviations is a challenge.

[0003] In the prior art, Snell et al. [J. Snell, K. Swersky, R. Zemel, Prototypical networks for few-shot learning, in: Advances in Neural Information Processing Systems (NeurIPS), 2017: 4080-4090.] proposed to take the average value of class features as a prototype, and segmented point clouds by the distance between point features and the prototype. However, this method does not take into account the complexity of the structure of point clouds itself, and the segmentation accuracy is not satisfactory enough. Zhao et al. [N. Zhao, T.-S. Chua, G. H. Lee, Few-shot 3D point cloud semantic segmentation, in: IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 8869-8878.] designed a point cloud semantic segmentation method based on multiple prototypes, which uses multiple prototypes to describe complex point cloud distributions, but its inference time is relatively long. SUMMARY

[0004] The present application provides a small sample point cloud semantic segmentation method, device, computer device, and storage medium, which aims to at least solve one of the above technical problems in the prior art to some extent.

[0005] To solve the above problems, the application provides the following technical solutions:

[0006] A small sample point cloud semantic segmentation method, comprising:

[0007] Obtain a support set point cloud and a query set point cloud corresponding to a mask, perform feature extraction on the support set point cloud and the query set point cloud to obtain support set features and query set features;

[0008] Calculate the distance between the support set features and the query set features, align the support set features according to the distance to obtain aligned support set features;

[0009] Obtain a corresponding prototype according to the mask of the aligned support set features, splice the prototype with the query set features to obtain expanded query set features, extract point cloud feature distribution information from the expanded query set features, correct the point cloud feature distribution information to obtain corrected point cloud feature distribution information, and segment the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information.

[0010] The technical solutions adopted by the embodiments of the application further include that the calculation of the distance between the support set features and the query set features is specifically:

[0011]

[0012] In the above formula, F s represents the support set features, F q represents the query set features, i represents the i-th feature in the support set features F s , j represents the j-th feature in the query set features F q , d i,j represents the distance between the i-th feature in the support set features F s and the j-th feature in the query set features F q .

[0013] The technical solutions adopted by the embodiments of the application further include that the alignment of the support set features according to the distance to obtain the aligned support set features is specifically:

[0014] The distance between the support set features F s and the query set features F q is inverted and normalized to obtain a weight weight:

[0015] weight i,j = softmax(-d i,j )

[0016] According to the weight weight, a feature inclined to a query set point cloud is obtained, and the feature is fused with a support set feature to obtain an aligned support set feature:

[0017]

[0018] The technical scheme adopted by the embodiment of the application further includes: extracting point cloud feature distribution information from the expanded query set feature, and correcting the point cloud feature distribution information to obtain corrected point cloud feature distribution information, and the specific process is as follows:

[0019] The point cloud feature distribution information F is extracted from the expanded query set feature by using a shared multi-layer perceptron e :

[0020] F e =MLP(Con(F q ,P))

[0021] In the above formula, Con represents a concatenation operation;

[0022] The point cloud feature distribution information F e is multiplied by the converted matrix to obtain a relationship matrix, and the point cloud feature distribution information F e is corrected by using the relationship matrix to obtain corrected point cloud feature distribution information

[0023]

[0024] The technical scheme adopted by the embodiment of the application further includes: the corrected prototype and the corrected query set feature are segmented from the corrected point cloud feature distribution information, and the specific process is as follows:

[0025]

[0026]

[0027] In the above formula, and respectively represent the segmented corrected prototype and the segmented corrected query set feature.

[0028] Another technical scheme adopted by the embodiment of the application is a small sample point cloud semantic segmentation device, which includes:

[0029] A feature extraction module is configured to obtain a support set point cloud and a query set point cloud corresponding to a mask, and extract features from the support set point cloud and the query set point cloud to obtain support set features and query set features.

[0030] Feature alignment module: used to calculate the distance between the support set features and the query set features, and align the support set features according to the distance to obtain the aligned support set features;

[0031] Distribution correction module: used to obtain the corresponding prototype based on the mask of the aligned support set features, concatenate the prototype with the query set features to obtain the expanded query set features, extract point cloud feature distribution information from the expanded query set features, correct the point cloud feature distribution information to obtain the corrected point cloud feature distribution information, and segment the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information.

[0032] The technical solution adopted in this application embodiment further includes: the feature alignment module calculates the distance between the support set features and the query set features, and aligns the support set features according to the distance, specifically as follows:

[0033]

[0034] In the above formula, F s F represents the support set features. q Let i represent the query set features, and let F represent the support set features. s The i-th feature in the query set, j represents the query set feature F. q The j-th feature in d i,j The support set feature F s The i-th feature and the query set feature F q The distance to the j-th feature;

[0035] For the support set features F s With query set features F q The distance is inverted and normalized to obtain the weight:

[0036] weight i,j =softmax(-d i,j )

[0037] Based on the weight, features that tend to favor the query set point cloud are obtained, and these features are fused with the support set features to obtain aligned support set features:

[0038]

[0039] Here, α is a hyperparameter used to preserve the original information of the features.

[0040] The technical solution adopted in this application embodiment further includes: the distribution correction module extracts point cloud feature distribution information from the expanded query set features, corrects the point cloud feature distribution information to obtain corrected point cloud feature distribution information, and segments the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information, specifically as follows:

[0041] Point cloud feature distribution information F is extracted from the augmented query set features using a shared multilayer perceptron. e :

[0042] F e =MLP(Con(F) q ,P))

[0043] In the above formula, Con represents the splicing operation;

[0044] The point cloud feature distribution information F e Multiplying the converted matrix by the relation matrix yields the relation matrix, which is then used to analyze the point cloud feature distribution information F. e The correction is performed to obtain the corrected point cloud feature distribution information.

[0045]

[0046] The corrected prototype and the corrected query set features are segmented from the corrected point cloud feature distribution information:

[0047]

[0048]

[0049] In the above formula, and These represent the corrected prototype and the corrected query set features, respectively. and This represents the final prototype and query set features, where β is a hyperparameter used to preserve the original information of the features.

[0050] Another technical solution adopted in this application embodiment is: a computer device, the computer device including a processor and a memory coupled to the processor, wherein,

[0051] The memory stores program instructions for implementing the small sample point cloud semantic segmentation method;

[0052] The processor is used to execute the program instructions stored in the memory to control the small sample point cloud semantic segmentation method.

[0053] Another technical solution adopted in this application embodiment is: a storage medium storing processor-executable program instructions, the program instructions being used to execute the small sample point cloud semantic segmentation method.

[0054] Compared to existing technologies, the beneficial effects of the embodiments of this application are as follows: The small sample point cloud semantic segmentation method, apparatus, computer equipment, and storage medium of the embodiments of this application extract features from the support set point cloud and query set point cloud using a feature extractor to obtain support set features and query set features. The feature alignment module performs alignment operations on the support set features and query set features, thereby reducing the inter-class distribution deviation caused by the difference between the support set point cloud and the query set point cloud. The distribution correction module adjusts the feature distribution information of the prototype and query set features, thereby reducing the intra-class distribution deviation caused by the difference in feature distribution between the prototype and query set features. While reducing inference time, the point cloud segmentation error is greatly reduced, and the accuracy of small sample point cloud semantic segmentation is improved. Attached Figure Description

[0055] Figure 1 This is a flowchart of a small-sample point cloud semantic segmentation method according to an embodiment of this application;

[0056] Figure 2 This is a schematic diagram of the framework of the small sample point cloud semantic segmentation system according to an embodiment of this application;

[0057] Figure 3 This is a schematic diagram of the feature alignment operation of the feature alignment module in the embodiments of this application;

[0058] Figure 4 This is a schematic diagram of the feature correction of the distribution correction module in an embodiment of this application;

[0059] Figure 5 This is a schematic diagram of the structure of a small sample point cloud semantic segmentation device according to an embodiment of this application;

[0060] Figure 6 This is a schematic diagram of the computer device structure according to an embodiment of this application;

[0061] Figure 7 This is a schematic diagram of the structure of the storage medium according to an embodiment of this application. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0063] The terms "first," "second," and "third" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationships and movements between components in a specific orientation (as shown in the figures). If the specific orientation changes, the directional indications also change accordingly. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or computer device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or computer devices.

[0064] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0065] Please see Figure 1 and Figure 2 , Figure 1 This is a flowchart of a few-sample point cloud semantic segmentation method according to an embodiment of this application. Figure 2 This is a schematic diagram of the framework of a few-sample point cloud semantic segmentation system according to an embodiment of this application. Specifically, the few-sample point cloud semantic segmentation method according to an embodiment of this application includes the following steps:

[0066] S100: Obtain the support set point cloud and the corresponding mask query set point cloud, and extract features from the support set point cloud and the query set point cloud using a feature extractor to obtain support set features and query set features.

[0067] S110: The feature alignment module calculates the distance between the support set features and the query set features, and aligns the support set features according to the distance to obtain the aligned support set features;

[0068] In this step, such as Figure 3 The diagram shown illustrates the feature alignment operation of the feature alignment module in this embodiment. Specifically, the feature alignment algorithm of the feature alignment module includes:

[0069] Step 1: Calculate the support set features F s With query set features F q Distance:

[0070]

[0071] In the above formula, F s F represents the support set features. q Let i represent the query set features, and let F represent the support set features. s The i-th feature in the query set, j represents the query set feature F. q The j-th feature in d i,j The support set feature F s The i-th feature and the query set feature F q The distance to the j-th feature.

[0072] The second step is to analyze the support set features F. s With query set features F q The distance is inverted and normalized to obtain the weight:

[0073] weight i,j =soft max(-d i,j (2)

[0074] The third step is to obtain the features that tend to favor the query set point cloud based on the weights, and then fuse these features with the support set features to obtain the aligned support set features:

[0075]

[0076] Here, α is a hyperparameter used to preserve the original information of the features.

[0077] It is understood that in the first stage, the feature alignment module is used to align the support set features and query set features, thereby reducing the inter-class distribution deviation caused by the difference between the support set point cloud and the query set point cloud.

[0078] S120: Obtain the corresponding prototype based on the mask of the aligned support set features, and use the distribution correction module to concatenate the prototype with the query set features to obtain the expanded query set features. Extract the point cloud feature distribution information from the expanded query set features, and correct the point cloud feature distribution information to obtain the corrected point cloud feature distribution information.

[0079] Please refer to the following in this step: Figure 4 This is a schematic diagram of feature correction of the distribution correction module in an embodiment of this application. Specifically, the feature correction algorithm of the distribution correction module includes:

[0080] S121: Combine prototype P with query set feature F q The features of the expanded query set are obtained by concatenating the features.

[0081] S122: Extract point cloud feature distribution information F from augmented query set features using a shared multilayer perceptron (MLP). e :

[0082] F e =MLP(Con(F) q ,P)) (4)

[0083] In formula (4), Con represents the splicing operation.

[0084] S123: Transfer point cloud feature distribution information F e Multiplying the matrix by the converted matrix yields the relation matrix, which is then used to analyze the point cloud feature distribution information F. e The correction is performed to obtain the corrected point cloud feature distribution information.

[0085]

[0086] S130: The corrected prototype and the corrected query set features are segmented from the corrected point cloud feature distribution information using the distribution correction module;

[0087] In this step, to preserve the original feature information, the distribution correction module adds the original feature information to both the corrected prototype and query set features during segmentation. The segmentation formula is as follows:

[0088]

[0089]

[0090] In the above formula, and These represent the corrected prototype and the corrected query set features, respectively. and This represents the final prototype and query set features, where β is a hyperparameter used to preserve the original information of the features.

[0091] It is understood that in the second stage, the distribution correction module of this application adjusts the feature distribution information, which can reduce intra-class bias and achieve better segmentation results.

[0092] Based on the above, the few-sample point cloud semantic segmentation method of this application extracts features from the support set point cloud and the query set point cloud using a feature extractor to obtain support set features and query set features. A feature alignment module is used to align the support set features and query set features, thereby reducing inter-class distribution deviations caused by differences between the support set point cloud and the query set point cloud. A distribution correction module is used to adjust the feature distribution information of the prototype and query set features, thereby reducing intra-class distribution deviations caused by differences in feature distribution between the prototype and query set features. This significantly reduces point cloud segmentation errors while lowering inference time, thus improving the accuracy of few-sample point cloud semantic segmentation.

[0093] Please see Figure 5 This is a schematic diagram of the structure of a small-sample point cloud semantic segmentation device according to an embodiment of this application. The small-sample point cloud semantic segmentation device 40 according to an embodiment of this application includes:

[0094] Feature extraction module 41: used to obtain the support set point cloud and the corresponding mask query set point cloud, and to extract features from the support set point cloud and the query set point cloud to obtain support set features and query set features;

[0095] Feature alignment module 42: used to calculate the distance between the support set features and the query set features, and align the support set features according to the distance to obtain the aligned support set features;

[0096] Distribution correction module 43: used to obtain the corresponding prototype according to the mask of the aligned support set features, concatenate the prototype with the query set features to obtain the expanded query set features, extract point cloud feature distribution information from the expanded query set features, correct the point cloud feature distribution information to obtain the corrected point cloud feature distribution information, and segment the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information.

[0097] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0098] The apparatus provided in this application can be applied to the foregoing method embodiments. For details, please refer to the description of the above method embodiments, which will not be repeated here.

[0099] Please see Figure 6 This is a schematic diagram of a computer device structure according to an embodiment of this application. The computer device 50 includes:

[0100] Memory 51 storing executable program instructions;

[0101] Processor 52 connected to memory 51;

[0102] The processor 52 is used to call the executable program instructions stored in the memory 51 and perform the following steps: obtain the support set point cloud and the corresponding query set point cloud; extract features from the support set point cloud and the query set point cloud to obtain support set features and query set features; calculate the distance between the support set features and the query set features; align the support set features according to the distance to obtain aligned support set features; obtain the corresponding prototype according to the mask of the aligned support set features; concatenate the prototype with the query set features to obtain expanded query set features; extract point cloud feature distribution information from the expanded query set features; correct the point cloud feature distribution information to obtain corrected point cloud feature distribution information; and segment the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information.

[0103] The processor 52 can also be referred to as a CPU (Central Processing Unit). The processor 52 may be an integrated circuit chip with signal processing capabilities. The processor 52 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor.

[0104] Please see Figure 7This is a schematic diagram of the structure of the storage medium in an embodiment of this application. The storage medium in this embodiment stores program instructions 61 capable of implementing the following steps: acquiring a support set point cloud and a query set point cloud with a corresponding mask; extracting features from the support set point cloud and the query set point cloud to obtain support set features and query set features; calculating the distance between the support set features and the query set features; aligning the support set features according to the distance to obtain aligned support set features; acquiring a corresponding prototype based on the mask of the aligned support set features; concatenating the prototype with the query set features to obtain expanded query set features; extracting point cloud feature distribution information from the expanded query set features; correcting the point cloud feature distribution information to obtain corrected point cloud feature distribution information; and segmenting the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information. The program instructions 61 can be stored in the aforementioned storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network computer device, etc.) or processor to execute all or part of the steps of the methods in various embodiments of this application. The aforementioned storage media include: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program instructions, or terminal computer devices such as computers, servers, mobile phones, and tablets. Servers can be standalone servers or cloud servers providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0106] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for semantic segmentation of point clouds with few samples, characterized in that, include: Obtain the support set point cloud and the corresponding mask query set point cloud, and perform feature extraction on the support set point cloud and the query set point cloud to obtain support set features and query set features. Calculate the distance between the support set features and the query set features, and align the support set features according to the distance to obtain the aligned support set features; Obtain the corresponding prototype based on the mask of the aligned support set features, concatenate the prototype with the query set features to obtain the expanded query set features, extract point cloud feature distribution information from the expanded query set features, and correct the point cloud feature distribution information to obtain the corrected point cloud feature distribution information. Segment the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information.

2. The small-sample point cloud semantic segmentation method according to claim 1, characterized in that, The calculation of the distance between the support set features and the query set features specifically involves: In the above formula, Indicates support set features, Indicates the characteristics of the query set. i Indicates support set features The first in i One characteristic, j Indicate query set characteristics The first in j One characteristic, Indicates support set features The Middle i Features and query set features The Middle j The distance to each feature.

3. The small-sample point cloud semantic segmentation method according to claim 2, characterized in that, The step of aligning the support set features according to the distance to obtain the aligned support set features is specifically as follows: For the support set features Features of the query set The distance is inverted and normalized to obtain the weights. : According to the weight The features of the query set point cloud are obtained, and these features are fused with the support set features to obtain the aligned support set features: in, It is a hyperparameter used to preserve the original information of the features.

4. The small-sample point cloud semantic segmentation method according to claim 3, characterized in that, The step of extracting point cloud feature distribution information from the expanded query set features and correcting the point cloud feature distribution information to obtain corrected point cloud feature distribution information is as follows: Point cloud feature distribution information is extracted from the augmented query set features using a shared multilayer perceptron. : In the above formula, Con Indicates a splicing operation; The point cloud feature distribution information Multiplying the matrix by the transposed matrix yields a relation matrix, which is then used to analyze the point cloud feature distribution information. The correction is performed to obtain the corrected point cloud feature distribution information. : 。 5. The small-sample point cloud semantic segmentation method according to claim 4, characterized in that, The specific steps for segmenting the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information are as follows: In the above formula, and These represent the corrected prototype and the corrected query set features, respectively. and This represents the final prototype and query set features. It is a hyperparameter used to preserve the original information of the features.

6. A small-sample point cloud semantic segmentation device, characterized in that, include: Feature extraction module: used to obtain the support set point cloud and the corresponding mask query set point cloud, and to extract features from the support set point cloud and the query set point cloud to obtain support set features and query set features; Feature alignment module: used to calculate the distance between the support set features and the query set features, and align the support set features according to the distance to obtain the aligned support set features; Distribution correction module: used to obtain the corresponding prototype based on the mask of the aligned support set features, concatenate the prototype with the query set features to obtain the expanded query set features, extract point cloud feature distribution information from the expanded query set features, correct the point cloud feature distribution information to obtain the corrected point cloud feature distribution information, and segment the corrected prototype and the corrected query set features from the corrected point cloud feature distribution information.

7. The small-sample point cloud semantic segmentation device according to claim 6, characterized in that, The feature alignment module calculates the distance between the support set features and the query set features, and aligns the support set features based on the distance as follows: In the above formula, Indicates support set features, Indicates the characteristics of the query set. i Indicates support set features The first in i One characteristic, j Indicate query set characteristics The first in j One characteristic, Indicates support set features The Middle i Features and query set features The Middle j Distance of each feature; For the support set features Features of the query set The distance is inverted and normalized to obtain the weights. : According to the weight The features of the query set point cloud are obtained, and these features are fused with the support set features to obtain the aligned support set features: 。 8. The small-sample point cloud semantic segmentation device according to claim 7, characterized in that, The distribution correction module extracts point cloud feature distribution information from the expanded query set features and corrects the point cloud feature distribution information to obtain corrected point cloud feature distribution information. Specifically, the corrected prototype and corrected query set features are segmented from the corrected point cloud feature distribution information as follows: Point cloud feature distribution information is extracted from the augmented query set features using a shared multilayer perceptron. : In the above formula, Con Indicates a splicing operation; The point cloud feature distribution information Multiplying the matrix by the transposed matrix yields a relation matrix, which is then used to analyze the point cloud feature distribution information. The correction is performed to obtain the corrected point cloud feature distribution information. : The corrected prototype and the corrected query set features are segmented from the corrected point cloud feature distribution information: In the above formula, and These represent the corrected prototype and the corrected query set features, respectively.

9. A computer device, characterized in that, The computer device includes a processor and a memory coupled to the processor, wherein, The memory stores program instructions for implementing the small sample point cloud semantic segmentation method according to any one of claims 1-5; The processor is used to execute the program instructions stored in the memory to control the small sample point cloud semantic segmentation method.

10. A storage medium, characterized in that, The device stores processor-executable program instructions for performing the small-sample point cloud semantic segmentation method according to any one of claims 1 to 5.

Images