CN112652070B - Three-dimensional model surface reduction method, device, equipment and medium - Google Patents
Three-dimensional model surface reduction method, device, equipment and medium Download PDFInfo
- Publication number
- CN112652070B CN112652070B CN202011519504.6A CN202011519504A CN112652070B CN 112652070 B CN112652070 B CN 112652070B CN 202011519504 A CN202011519504 A CN 202011519504A CN 112652070 B CN112652070 B CN 112652070B
- Authority
- CN
- China
- Prior art keywords
- reduction
- value
- plane
- dimensional model
- face
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention relates to the technical field of three-dimensional models, and discloses a method, a device, equipment and a medium for reducing the surface of a three-dimensional model, wherein the method comprises the following steps: constructing three-dimensional model data corresponding to the three-dimensional file to be processed by acquiring the three-dimensional file to be processed; performing surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result; when the face reduction result is that face reduction is needed, determining a target face numerical value corresponding to each sub-model in the three-dimensional model data according to the reference face reduction value; according to all target surface values corresponding to the sub-models, corresponding surface reduction is carried out on the three-dimensional model data; and compressing the three-dimensional model data subjected to the surface reduction into a three-dimensional file with the surface reduction and outputting the three-dimensional file. The method and the device realize automatic identification of the face reduction result and the reference face reduction value, automatic output of the target face numerical value, corresponding face reduction according to the target face numerical value, and compressed output, do not need manual operation, improve the face reduction efficiency of the three-dimensional model, and improve the face reduction quality of the three-dimensional model.
Description
Technical Field
The invention relates to the technical field of three-dimensional models, in particular to a surface reduction method, a surface reduction device, surface reduction equipment and a surface reduction medium of a three-dimensional model.
Background
Three-dimensional models have been widely used in various fields, such as the home industry, the construction industry, the medical industry, the film industry, the video game industry, and in scientific research and engineering applications. The three-dimensional model object is composed of triangular patches, and the rendering and manufacturing of the three-dimensional model have extremely high requirements on a hardware system and modeling software. Generally, the number of faces of a three-dimensional model ranges from tens to millions, a three-dimensional model with slightly complex hardware configuration is subjected to different degrees of clamping, and the rendering time of the model also changes along with the complexity of the model and the number of faces of the model, so that the number of triangular faces of the three-dimensional model is a key factor influencing the rendering efficiency and the number of frames of the model.
At present, the prior art mainly adopts the following two methods to reduce the surface of the three-dimensional model: the first is to use some related face-reducing plug-ins officially provided by mainstream modeling software to automatically reduce the face to some extent; the second is to reconstruct a simplified model according to the contour of the original model, or to manually reduce the surface on the basis of the original model.
However, although the above two methods can achieve the face reduction of the three-dimensional model, they have different defects. Specifically, plug-ins used in the first method generally need to pay, and the result generated by the polygonal face reduction tool cannot meet special requirements, so that the model characteristics are difficult to finely process, and the triangular face patch is reduced to a certain proportion, so that the triangular face patch is deformed or deformed to different degrees; the second method needs human eyes to judge the face reduction effect, and although the method has good effect and the number of the face slices can be manually controlled, the method is time-consuming, labor-consuming and extremely high in labor cost and cannot adapt to large-scale model processing.
Therefore, it is necessary to provide a technical solution to solve the above technical problems.
Disclosure of Invention
The invention provides a method and a device for reducing the surface of a three-dimensional model, computer equipment and a storage medium, which realize the automatic completion of the surface reduction operation of the three-dimensional model, do not need manual operation, reduce the labor cost, improve the surface reduction efficiency of the three-dimensional model, ensure the surface reduction effectiveness and no distortion of the three-dimensional model and improve the surface reduction quality of the three-dimensional model.
A method of face reduction of a three-dimensional model, comprising:
acquiring a three-dimensional file to be processed, and constructing three-dimensional model data corresponding to the three-dimensional file to be processed;
performing surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result;
when the face reduction result is the face reduction required, determining a target face numerical value corresponding to each sub model in the three-dimensional model data according to the reference face reduction value;
according to all target surface values corresponding to the sub-models, corresponding surface reduction is carried out on the three-dimensional model data;
compressing the three-dimensional model data subjected to surface reduction into a surface reduction three-dimensional file and outputting the surface reduction three-dimensional file;
the subtracting the three-dimensional model data to obtain a subtracting result and a reference subtracting value corresponding to the subtracting result includes:
carrying out plane data collection on the three-dimensional model data to obtain a plurality of plane data; the plane data comprises a plane image, a plane number and a plane size;
inputting all the plane images into a minus surface recognition model, extracting texture features of all the plane images through the minus surface recognition model, and obtaining the minus surface result recognized by the minus surface recognition model according to the extracted texture features;
when the face reduction result is that face reduction is needed, vector conversion is carried out on the number of the plane faces and the plane size in each plane data to obtain a vector matrix corresponding to the three-dimensional model data;
and inputting the vector matrix into a reference recognition model, and performing reference surface reduction prediction on the vector matrix through the reference recognition model to obtain the reference surface reduction value corresponding to the surface reduction result.
A face reducing apparatus for a three-dimensional model, comprising:
the acquisition module is used for acquiring a three-dimensional file to be processed and constructing three-dimensional model data corresponding to the three-dimensional file to be processed;
the identification module is used for carrying out surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result;
the determining module is used for determining a target surface numerical value corresponding to each sub-model in the three-dimensional model data according to the reference surface reduction value when the surface reduction result is the surface reduction required;
the face reduction module is used for carrying out corresponding face reduction on the three-dimensional model data according to all target face numerical values corresponding to the sub models;
the output module is used for compressing the three-dimensional model data subjected to surface reduction into a surface-reduced three-dimensional file and outputting the surface-reduced three-dimensional file;
the identification module is further configured to:
carrying out plane data collection on the three-dimensional model data to obtain a plurality of plane data; the plane data comprises a plane image, a plane number and a plane size;
inputting all the plane images into a minus surface recognition model, extracting texture features of all the plane images through the minus surface recognition model, and obtaining the minus surface result recognized by the minus surface recognition model according to the extracted texture features;
when the face reduction result is that face reduction is needed, vector conversion is carried out on the number of the plane faces and the plane size in each plane data to obtain a vector matrix corresponding to the three-dimensional model data;
and inputting the vector matrix into a reference recognition model, and performing reference reduced surface prediction on the vector matrix through the reference recognition model to obtain the reference reduced surface value corresponding to the reduced surface result.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the above-described method of facelift of a three-dimensional model when executing the computer program.
A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method of facelift of a three-dimensional model.
According to the three-dimensional model surface reduction method and device, the computer equipment and the storage medium, the three-dimensional model data corresponding to the three-dimensional file to be processed is constructed by acquiring the three-dimensional file to be processed; performing surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result; when the face reduction result is that face reduction is needed, determining a target face numerical value corresponding to each sub-model in the three-dimensional model data according to the reference face reduction value; according to all target surface values corresponding to the sub-models, corresponding surface reduction is carried out on the three-dimensional model data; the three-dimensional model data after surface reduction is compressed into a surface reduction three-dimensional file and output, so that the object surface value of each sub-model in the three-dimensional model data is automatically output by constructing the three-dimensional model data, the surface reduction result and the reference surface reduction value are automatically identified, when the surface reduction result is required to be reduced, the corresponding surface reduction is carried out on each sub-model surface reduction according to the object surface value corresponding to each sub-model surface reduction, and the surface reduction three-dimensional file is compressed and output, thereby automatically completing the surface reduction operation of the three-dimensional model, avoiding manual operation, reducing the labor cost, improving the surface reduction efficiency of the three-dimensional model, ensuring the effectiveness and undistortion of the surface reduction of the three-dimensional model, and improving the surface reduction quality of the three-dimensional model.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic diagram of an application environment of a three-dimensional model surface reduction method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for face reduction of a three-dimensional model in an embodiment of the invention;
FIG. 3 is a flowchart of step S20 of a method for surface reduction of a three-dimensional model according to an embodiment of the present invention;
FIG. 4 is a flowchart of step S20 of a method for face subtraction of a three-dimensional model according to another embodiment of the present invention;
FIG. 5 is a flowchart of step S30 of a method for face subtraction of a three-dimensional model according to an embodiment of the present invention;
FIG. 6 is a functional block diagram of a transformation module of the face reduction apparatus for a three-dimensional model according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a computer device in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
The method for reducing the surface of the three-dimensional model provided by the invention can be applied to an application environment such as that shown in figure 1, wherein a client (computer device) is communicated with a server through a network. The client (computer device) includes, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server may be implemented as a stand-alone server or as a server cluster comprised of multiple servers.
In an embodiment, as shown in fig. 2, a method for reducing a surface of a three-dimensional model is provided, which mainly includes the following steps S10 to S50:
s10, acquiring a three-dimensional file to be processed, and constructing three-dimensional model data corresponding to the three-dimensional file to be processed.
Understandably, the three-dimensional file to be processed is a file of an original three-dimensional model which is not processed and needs to be subjected to surface reduction processing, the three-dimensional model data is constructed in a process of opening the three-dimensional file to be processed through application software, for example, the application software is 3dsMax software, the three-dimensional model data can be constructed through the 3dsMax software, the three-dimensional model data is data displayed by decoding through the application software, the three-dimensional model data comprises a three-dimensional model and sub models under the three-dimensional model, the three-dimensional model is a model with a three-dimensional structure, and the sub models are models of sub objects forming the three-dimensional model.
And S20, performing surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result.
Understandably, the processing mode of the face reduction identification can be set according to requirements, whether the three-dimensional model needs to be subjected to face reduction can be identified through the processing process of the face reduction identification, i.e., whether the three-dimensional model data requires a face reduction process, and, in the case of a face reduction, a reference face reduction value is identified, setting the reference face reduction value to zero in the case of face reduction, indicating that the three-dimensional model data does not need face reduction operation, indicating that the three-dimensional model data has reached the face reduction impossible case, if the rendering distortion or the three-dimensional model damage result can occur after the surface reduction operation, the surface reduction result shows the result of whether the surface reduction operation is required to be performed on the three-dimensional model data, the reference face subtraction value is a reference value of a face subtraction determined for the three-dimensional model, for example, the processing manner of face subtraction recognition may be to count the number of faces of the three-dimensional model data to obtain the total number of faces of the three-dimensional model data, determine whether face subtraction is required according to the total number of faces, if the face reduction is needed, determining a reference face reduction value by combining the volume value, the surface area value and the size value of each sub model, the processing mode of the face subtraction recognition can also be a plane data collection for the three-dimensional model data to obtain a plurality of plane data, extracting texture features of the plane images in all the plane data through a face reduction recognition model to obtain an output face reduction result, under the condition of surface reduction, the number of plane surfaces and the plane size in each plane data are subjected to vector conversion into a vector matrix, and performing reference surface reduction prediction on the vector matrix through a reference identification model, and predicting the processing process of the reference surface reduction value.
The volume value is a value of a volume of the sub-model, the surface area value is a value of a surface area of the sub-model, the size value is a value of a length and a width of the sub-model, the plane data is data obtained by capturing six directions, namely, an upper direction, a lower direction, a left direction, a right direction, a front direction and a rear direction, of a space body of the three-dimensional model, the plane data comprises a plane image, a plane surface number and a plane size, the plane image is a captured binary image, namely an image after graying processing, the plane surface number is all the surface numbers in the plane image captured in one direction, the plane size is an extreme length and an extreme width of the three-dimensional model in one direction, the extreme length is a maximum length of the three-dimensional model in capturing, the extreme width is a maximum width of the three-dimensional model in capturing, and the plane size comprises the extreme length and the extreme width.
In an embodiment, after the step S20, that is, after the performing the face reduction recognition on the three-dimensional model to obtain a face reduction result and a reference face reduction value corresponding to the face reduction result, the method further includes: and S60, when the face reduction result shows that face reduction is not needed, compressing the three-dimensional model data into a face-reduced three-dimensional file and outputting the face-reduced three-dimensional file.
Understandably, under the condition that the three-dimensional model does not need surface reduction, the three-dimensional model data is directly compressed into the surface reduction three-dimensional file, namely, the three-dimensional model data is compressed by using an LZMA (Lempel-Ziv-Markov chain-Algorithm) compression Algorithm to obtain the surface reduction three-dimensional file, wherein the LZMA compression Algorithm uses an improved compression Algorithm of LZ77 (lossless compression Algorithm) supported by interval coding and an Algorithm of a preprocessing program special for binary system.
According to the method and the device, when the face reduction result is automatically judged that face reduction is not needed, three-dimensional model data are automatically compressed to obtain a face reduction three-dimensional file, the cost of manual automatic identification and compression is reduced, the three-dimensional file to be processed is automatically compressed, and the size of the file is reduced.
In an embodiment, as shown in fig. 3, the step S20 of performing the face reduction recognition on the three-dimensional model to obtain a face reduction result and a reference face reduction value corresponding to the face reduction result includes:
s201, carrying out surface number statistics on the three-dimensional model data to obtain the total surface number of the three-dimensional model data.
Understandably, the surface number statistics is a process of performing statistics on the surface number of each sub-model in the three-dimensional model, and the total surface number is a sum of the surface numbers obtained after the surface number statistics, that is, how many polygons are in the three-dimensional model.
S202, judging whether the total number of faces is larger than a preset number of faces threshold value.
Understandably, the preset face number threshold is a preset face number value, the preset face number threshold is a set face number value from face reduction to minimum, for example, the preset face number threshold is 8, 10, etc., and whether the total face number is smaller than the preset face number threshold is determined, if the total face number is smaller than the preset face number threshold, it indicates that the face reduction can not be performed, at this time, the face reduction is not required.
S203, if the total number of the surfaces is larger than the preset number of the surfaces threshold, determining the surface reduction result as the surface to be reduced.
Understandably, if the total number of faces is greater than the preset number of faces threshold, it indicates that the three-dimensional model data needs face subtraction, that is, face subtraction is needed for a sub-model in the three-dimensional model, at this time, the face subtraction result is determined as a face subtraction needed, and the face subtraction result includes a face subtraction needed and a face subtraction not needed.
S204, obtaining the volume value, the surface area value and the size value of each sub-model in the three-dimensional model data.
Understandably, the volume value, the surface area value and the size value of each sub model in the three-dimensional model can be obtained from the constructed three-dimensional model data.
S205, sequencing the volume values of all the submodels to obtain a first sequencing result, sequencing the surface area values of all the submodels to obtain a second sequencing result, and sequencing the size values of all the submodels to obtain a third sequencing result;
understandably, the volume values of the sub-models are sorted in descending order, all the sorted volume values are determined as the first sorting result, the surface area values of the sub-models are sorted in descending order, all the sorted surface area values are determined as the second sorting result, the size values of the sub-models are sorted in descending order, and all the sorted size values are determined as the third sorting result.
S206, determining the reference face reduction value corresponding to the face reduction result according to the first sorting result, the second sorting result and the third sorting result.
Understandably, the first volume value of the first third of the first sorting result and the last three of the volume value of the first sorting result that are next to the last sequence are averaged to obtain a first sorting average value, the second sorting result and the third sorting result are processed in the same way to obtain a second sorting average value and a third sorting average value, the first sorting average value, the second sorting average value and the third sorting average value are mapped and converted into values of the same dimension (i.e., a reference decreasing dimension), that is, a range of values in which the first sorting average value falls is detected, and then the values are mapped into a first reference value corresponding to the range of values, the second sorting average value and the third sorting average value are mapped and converted in the same way to obtain a second reference value and a third reference value, the first reference value, the second reference value and the third reference value are input into a reference decreasing function, and the reference decreasing value is calculated by the reference decreasing function, wherein the reference decreasing function is:
L＝α 1 L 1 +α 2 L 2 +α 3 L 3
wherein:
l is a reference reduced value;
L 1 is a first reference value;
L 2 is a second reference value;
L 3 is a third reference value;
α 1 is a preset first reference weight;
α 2 is a preset second reference weight;
α 3 is a preset thirdThe basis weight.
The invention realizes the surface number statistics of the three-dimensional model data to obtain the total surface number of the three-dimensional model data; judging whether the total number of faces is larger than a preset number of faces threshold value; if the total number of the surfaces is larger than the preset surface number threshold, determining the surface reduction result as the surface to be reduced; obtaining the volume value, the surface area value and the size value of each sub-model in the three-dimensional model data; sequencing the volume values of all the submodels to obtain a first sequencing result, sequencing the surface area values of all the submodels to obtain a second sequencing result, and sequencing the size values of all the submodels to obtain a third sequencing result; and determining the reference face reduction value corresponding to the face reduction result according to the first sequencing result, the second sequencing result and the third sequencing result, so that whether face reduction is required or not is automatically judged through face number statistics and preset face number threshold values, and the reference face reduction value is automatically determined by combining the sequencing results through sequencing the volume value, the surface area value and the size value of each submodel in the three-dimensional model data, so that manual operation is reduced, manual subjective identification is not required, the identification accuracy and reliability are improved, and the quality of face reduction of the three-dimensional model is improved.
In an embodiment, as shown in fig. 4, in the step S20, that is, performing surface reduction recognition on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result, the method further includes:
s207, carrying out plane data collection on the three-dimensional model data to obtain a plurality of plane data; the plane data includes a plane image, a plane number and a plane size.
Understandably, the plane data collection is to capture six planes of the three-dimensional model data to obtain the plane images of the three-dimensional model data in the six planes; and a processing procedure of summarizing the number of planes of each plane image to obtain the number of planes corresponding to each plane image, and measuring the size of each plane image to obtain the size of each plane corresponding to each plane image.
In an embodiment, in step S207, performing plane data collection on the three-dimensional model data to obtain a plurality of plane data; the plane data includes a plane image, a plane number and a plane size, including:
s2071, capturing six surfaces of the three-dimensional model data to obtain the plane images of the three-dimensional model data on six planes.
Understandably, the capturing process is a process of capturing the planar image from six directions of the upper, lower, left, right, front, and rear of the spatial form by the application software, and may also be understood as a process of capturing an image from six directions of the upper, lower, left, right, front, and rear and then binarizing (i.e., graying) the captured image, wherein the binarizing (i.e., graying) is a process of converting a color image including brightness and color into a grayscale image.
And S2072, summarizing the number of planes of each plane image to obtain the number of planes corresponding to each plane image, and measuring the size of each plane image to obtain the plane size corresponding to each plane image.
Understandably, the surface numbers are summarized to calculate the number of the surfaces in the plane image, so as to obtain the plane surface number of the plane image, and the size measurement is a process of measuring the limit length and the limit width of the plane dimension in the plane image.
S2073, determining one plane image, the number of plane surfaces corresponding to the plane image, and the plane size as one plane data.
Understandably, the plane image of one orientation and the plane surface number and the plane size corresponding to the plane image are marked as the plane data of the orientation.
S208, inputting all the plane images into a minus surface recognition model, extracting texture features of all the plane images through the minus surface recognition model, and obtaining the minus surface result recognized by the minus surface recognition model according to the extracted texture features.
Understandably, the face reduction recognition model is a trained deep neural network model, a network structure of the face reduction recognition model can be set according to requirements, preferably, the network structure of the face reduction recognition model is a network structure of VGG19, the face reduction recognition model is a model for extracting the texture features in all the plane images and recognizing a face reduction result whether the face reduction is required, the texture features are features related to length, density or superposition of sides of polygons in the plane images, and as the number of plane faces in the plane images is larger, the texture features are more obvious, the face reduction result whether the face reduction is required can be more accurately determined by extracting the texture features, the texture features in each plane image extracted by the face reduction recognition model are combined, and the face reduction result is recognized by integrating the texture features in each plane image.
In an embodiment, before the step S208, that is, before all the plane images are input into the minus plane recognition model, the method includes:
s2081, obtaining a training sample set; the training sample set comprises a plurality of training samples; each of the training samples is associated with a minus label; the face reduction label comprises a face needing to be reduced and a face needing not to be reduced; one of the training samples includes six planar sample images of one three-dimensional model sample collected historically.
Understandably, the training sample set is a set of the training samples, the training samples are three-dimensional model samples collected historically, the three-dimensional model samples are data with three-dimensional models as samples, the training samples include six plane sample images in the three-dimensional model samples, the face reduction label indicates a result of whether the corresponding training sample needs face reduction, that is, the face reduction label is a label given by whether the historical training sample needs face reduction or not, the face reduction label includes face reduction needing and face reduction not needing, and one training sample is associated with one face reduction label.
S2082, inputting the training sample into a deep neural network model containing initial parameters.
Understandably, the deep neural network model includes the initial parameters.
S2083, extracting the texture features of all the plane sample images in the training sample through the deep neural network model.
S2084, obtaining a face reduction sample result output by the deep neural network model according to the extracted texture features, and determining a loss value according to the face reduction sample result and the matching degree of the face reduction label;
understandably, the deep neural network model identifies according to the extracted textural features, identifies whether the training sample needs to be reduced, outputs an identification result as a result of the reduced surface sample, the reduced surface sample result comprises a reduced surface sample and a reduced surface sample, the reduced surface sample result indicates whether the training sample needs to be reduced, matches the reduced surface sample result corresponding to the training sample with the reduced surface label, and can match the difference degree between the reduced surface sample result and the reduced surface label by using a binary cross entropy loss algorithm, thereby calculating the loss value, measuring the difference between the reduced surface sample result and the reduced surface label, and continuously enabling the deep neural network model to approach to an accurate identification result through the loss value, thereby improving the identification accuracy.
S2085, when the loss value does not reach a preset convergence condition, iteratively updating the initial parameters of the deep neural network model until the loss value reaches the preset convergence condition, and recording the converged deep neural network model as a trained face-reducing recognition model.
Understandably, the convergence condition may be a condition that the loss value is small and does not decrease any more after the loss value is calculated for 1000 times, that is, when the loss value is small and does not decrease any more after the loss value is calculated for 1000 times, the training is stopped, and the deep neural network model after convergence is recorded as a trained reduced surface recognition model; the convergence condition may also be a condition that the loss value is smaller than a set threshold, that is, when the loss value is smaller than the set threshold, the training is stopped, and the converged deep neural network model is recorded as a reduced surface recognition model after the training is completed, so that when the loss value does not reach the preset convergence condition, the initial parameters of the deep neural network model are continuously adjusted, and the step of extracting the texture features of all the plane sample images in the training sample through the deep neural network model is triggered, so that accurate results can be continuously drawn together, and the recognition accuracy is increased. Therefore, the face reduction recognition can be optimized, and the accuracy and the reliability of the face reduction recognition of the three-dimensional model are improved.
The invention realizes the purpose of obtaining the training sample set; extracting the texture features of all the plane sample images in the training sample through the deep neural network model; obtaining a face reduction sample result output by the deep neural network model according to the extracted texture features, and determining a loss value according to the matching degree of the face reduction sample result and the face reduction label; when the loss value does not reach the preset convergence condition, the initial parameters of the deep neural network model are updated in an iterative mode until the loss value reaches the preset convergence condition, the converged deep neural network model is recorded as a training-finished face-reducing recognition model, and therefore the extraction of texture features is carried out through historical training samples, the face-reducing sample result is recognized, the initial parameters are iterated continuously according to the loss value, the result that whether the input image needs to be subjected to face reduction or not is recognized accurately and quickly, the accuracy and quality of face-reducing recognition of the three-dimensional model can be improved, cost is reduced, and training efficiency is improved.
S209, when the face subtraction result is that face subtraction is needed, vector conversion is carried out on the number of the plane faces and the plane size in each plane data, and a vector matrix corresponding to the three-dimensional model data is obtained.
Understandably, if the subtraction result is that subtraction is required, vector conversion is performed on the plane number and the plane size in each piece of plane data, the vector is converted into a binary code which converts the plane number and the plane size into the same dimension (preset dimension number) corresponding to the plane number and the plane size, for example, all the plane numbers and the plane size converted values in one piece of plane data are converted into a one-dimensional array, all the plane data are subjected to vector conversion and then combined into a multi-dimensional array, and the multi-dimensional array is determined as the vector matrix.
And S210, inputting the vector matrix into a reference identification model, and performing reference reduced surface prediction on the vector matrix through the reference identification model to obtain the reference reduced surface value corresponding to the reduced surface result.
Understandably, the reference recognition model is a trained clustering model, the reference recognition model performs clustering through vector matrixes collected historically, a hidden mapping relation between a reference reduced surface value and the vector matrixes can be learned and predicted, and the reference reduced surface value corresponding to the vector matrixes can be predicted by performing reference reduced surface prediction on the vector matrixes through the reference recognition model.
According to the invention, the three-dimensional model data is subjected to plane data collection to obtain a plurality of plane data; inputting all the plane images into a minus plane recognition model, extracting texture features of all the plane images through the minus plane recognition model, and acquiring the minus plane result recognized by the minus plane recognition model according to the extracted texture features; when the face reduction result is that face reduction is needed, vector conversion is carried out on the number of the plane faces and the plane size in each plane data to obtain a vector matrix corresponding to the three-dimensional model data; the vector matrix is input into a reference recognition model, reference face reduction prediction is carried out on the vector matrix through the reference recognition model, and a reference face reduction value corresponding to a face reduction result is obtained, so that plane data collection is carried out on three-dimensional model data, texture features are extracted through the face reduction recognition model, whether the three-dimensional model data need a face reduction result or not can be recognized, and the reference face reduction value can be predicted through vector conversion and the reference recognition model, therefore, whether the face reduction result needs face reduction or not can be accurately and scientifically automatically recognized, the reference face reduction value can be rapidly predicted, a data basis is provided for face reduction of a subsequent three-dimensional model, and recognition accuracy and quality of face reduction of the three-dimensional model are improved.
And S30, when the face reduction result is that the face needs to be reduced, determining a target face numerical value corresponding to each sub-model in the three-dimensional model data according to the reference face reduction value.
Understandably, when the face reduction result is that face reduction is required, the target face value of each sub-model in the three-dimensional model data is calculated by combining the reference face reduction value and the volume value, the volume ratio value, the surface area ratio value, the size value and the size ratio value of each sub-model in the three-dimensional model data, and the target face value of one sub-model indicates the value of the number of faces which the sub-model needs face reduction to reach.
In an embodiment, as shown in fig. 5, the determining, in step S30, a target surface value corresponding to each sub-model in the three-dimensional model data according to the reference reduced surface value includes:
s301, obtaining the volume value, the volume ratio, the surface area value, the surface area ratio, the size value and the size ratio of each sub-model in the three-dimensional model data.
Understandably, the three-dimensional model data further includes the volume fraction value, the surface fraction value and the dimension fraction value of each sub-model, the volume fraction value is a percentage of the volume value of the sub-model to the total volume of the three-dimensional model, the surface fraction value is a percentage of the surface value of the sub-model to the total surface area of the three-dimensional model, and the dimension fraction value is a percentage of the dimension value of the sub-model to the total dimension of the three-dimensional model, that is, a percentage of the product of the limit length and the limit width of each sub-model to the product of the length and the width of the three-dimensional model.
S302, according to the volume value, the surface area value and the size value of each sub-model, a protection surface value corresponding to each sub-model is determined.
Understandably, according to the volume value, the surface area value and the size value of the sub-model, the protection surface value corresponding to the sub-model can be mapped, the protection surface value is the number of surfaces which ensure that the outline of the sub-model is not damaged by the subtractive surface, namely, which section the volume value, the surface area value and the size value respectively fall into, and then a protection surface value is mapped according to the three sections respectively falling into, wherein the sections of three dimensions correspond to a protection surface value mapping.
And S303, multiplying the reference reduction value by the volume ratio of each submodel to obtain a first ratio corresponding to each submodel, multiplying the reference reduction value by the surface ratio of each submodel to obtain a second ratio corresponding to each submodel, and multiplying the reference reduction value by the size ratio of each submodel to obtain a third ratio corresponding to each submodel.
S304, determining the target face value corresponding to the sub-model according to the protection face value, the first ratio, the second ratio and the third ratio corresponding to the same sub-model.
Understandably, the estimated surface value corresponding to the submodel is determined through the first, second and third occupation ratios corresponding to the submodel, and the target surface value corresponding to the submodel is determined through comparing the estimated surface value corresponding to the submodel with the protection surface value.
The invention realizes the purpose that the volume value, the volume ratio, the surface area value, the surface area ratio, the size value and the size ratio of each submodel in the three-dimensional model data are obtained; determining a protection surface numerical value corresponding to each sub-model according to the volume value, the surface area value and the size value of each sub-model; multiplying the reference reduction value by the volume ratio of each submodel to obtain a first ratio corresponding to each submodel, multiplying the reference reduction value by the surface area ratio of each submodel to obtain a second ratio corresponding to each submodel, and multiplying the reference reduction value by the size ratio of each submodel to obtain a third ratio corresponding to each submodel; and determining the target face value corresponding to the sub-model according to the protection face value, the first ratio, the second ratio and the third ratio corresponding to the same sub-model, so that the target face value corresponding to each sub-model is scientifically and accurately determined automatically, the face reduction quality of each sub-model can be ensured, the face reduction target face value of each sub-model is provided, the accuracy is improved for subsequent face reduction, and the distortion and the damage of the face reduction are avoided.
In an embodiment, the determining, in step S304, the target surface value corresponding to the sub-model according to the protection surface value, the first ratio value, the second ratio value, and the third ratio value corresponding to the same sub-model includes:
s3041, rounding the largest value of the first, second, and third ratios corresponding to the submodel, and determining the rounded value as an estimated surface value corresponding to the submodel.
Understandably, the rounding is to round the largest value among the first ratio, the second ratio and the third ratio downwards, the rounded value is determined as the estimated surface value corresponding to the submodel, and the estimated surface value is a value of the surface number which is estimated to be reached by surface reduction of the submodel.
S3042, comparing the estimated surface value corresponding to the sub-model with the protection surface value.
S3043, if the estimated surface value corresponding to the sub-model is smaller than or equal to the protection surface value, determining the protection surface value as the target surface value corresponding to the sub-model.
S3044, if the estimated surface value corresponding to the sub-model is larger than the protection surface value, determining the estimated surface value as the target surface value corresponding to the sub-model.
The invention realizes that the largest value in the first, second and third ratio values corresponding to the submodel is rounded and determined as the pre-estimated surface value corresponding to the submodel; comparing the estimated surface value corresponding to the sub-model with the protection surface value; if the estimated surface value corresponding to the sub-model is smaller than or equal to the protection surface value, determining the protection surface value as the target surface value corresponding to the sub-model; and if the estimated surface value corresponding to the sub-model is greater than the protection surface value, determining the estimated surface value as the target surface value corresponding to the sub-model, thus providing a method for determining the target surface value of each sub-model, providing a target for surface reduction of a subsequent three-dimensional model, and improving the surface reduction quality of the three-dimensional model.
And S40, performing corresponding surface reduction on the three-dimensional model data according to all the target surface numerical values corresponding to the sub-models.
Understandably, the application program subtracts the surface of each submodel according to the target surface value of each submodel, so that corresponding surface subtraction operation can be automatically carried out, display surfaces which do not accord with rules and are unnecessary are subtracted, the surface number reserved by each submodel is close to the target surface value corresponding to each submodel until the target surface value is reached, the corresponding surface subtraction operation is finished, and finally, the three-dimensional model data after surface subtraction is obtained after the corresponding surface subtraction operation of all submodels is finished.
And S50, compressing the three-dimensional model data subjected to surface reduction into a three-dimensional file with reduced surface and outputting the three-dimensional file.
Understandably, compressing the three-dimensional model data after surface reduction into the three-dimensional file with the surface reduced, namely applying an LZMA (Lempel-Ziv-Markov chain-Algorithm) compression Algorithm to compress the three-dimensional model data after surface reduction to obtain the three-dimensional file with the surface reduced, wherein the LZMA compression Algorithm uses an improved compression Algorithm of LZ77 (lossless compression Algorithm) supported by interval coding and an Algorithm of a special binary preprocessing program to output the three-dimensional file with the surface reduced to finish the surface reduction process of the three-dimensional file to be processed, the three-dimensional file with the surface reduced guarantees the subsequent rendering effect through experimental data, and the capacity of the three-dimensional file with the surface reduced is less than that of the three-dimensional file to be processed by more than 10%.
The method and the device realize that the three-dimensional model data corresponding to the three-dimensional file to be processed is constructed by acquiring the three-dimensional file to be processed; performing surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result; when the face reduction result is that face reduction is needed, determining a target face numerical value corresponding to each sub-model in the three-dimensional model data according to the reference face reduction value; according to all target surface values corresponding to the sub-models, corresponding surface reduction is carried out on the three-dimensional model data; the three-dimensional model data after surface reduction is compressed into a surface reduction three-dimensional file and output, so that the object surface value of each sub-model in the three-dimensional model data is automatically output by constructing the three-dimensional model data, the surface reduction result and the reference surface reduction value are automatically identified, when the surface reduction result is required to be reduced, the corresponding surface reduction is carried out on each sub-model surface reduction according to the object surface value corresponding to each sub-model surface reduction, and the surface reduction three-dimensional file is compressed and output, thereby automatically completing the surface reduction operation of the three-dimensional model, avoiding manual operation, reducing the labor cost, improving the surface reduction efficiency of the three-dimensional model, ensuring the effectiveness and undistortion of the surface reduction of the three-dimensional model, and improving the surface reduction quality of the three-dimensional model.
In an embodiment, a surface reduction device of a three-dimensional model is provided, and the surface reduction device of the three-dimensional model corresponds to the surface reduction method of the three-dimensional model in the above embodiment one to one. As shown in fig. 6, the surface reducing device of the three-dimensional model includes an obtaining module 11, a recognition module 12, a determination module 13, a surface reducing module 14 and an output module 15. The functional modules are explained in detail as follows:
the acquisition module 11 is configured to acquire a three-dimensional file to be processed and construct three-dimensional model data corresponding to the three-dimensional file to be processed;
the identification module 12 is configured to perform face subtraction identification on the three-dimensional model data to obtain a face subtraction result and a reference face subtraction value corresponding to the face subtraction result;
a determining module 13, configured to determine, according to the reference face reduction value, a target face numerical value corresponding to each sub-model in the three-dimensional model data when the face reduction result is that face reduction is required;
the face reduction module 14 is used for performing corresponding face reduction on the three-dimensional model data according to all target face numerical values corresponding to the sub-models;
and the output module 15 is used for compressing the three-dimensional model data subjected to the surface reduction into a surface-reduced three-dimensional file and outputting the surface-reduced three-dimensional file.
For the specific definition of the surface reduction device of the three-dimensional model, reference may be made to the above definition of the surface reduction method of the three-dimensional model, and details are not repeated here. The modules in the surface reduction device of the three-dimensional model can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of face subtraction of a three-dimensional model.
In one embodiment, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the processor implements the method for reducing a three-dimensional model in the above embodiments.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, implements the method for faceting a three-dimensional model in the above-described embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above-mentioned embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein.
Claims (9)
1. A method for reducing surface area of a three-dimensional model, comprising:
acquiring a three-dimensional file to be processed, and constructing three-dimensional model data corresponding to the three-dimensional file to be processed;
performing surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result;
when the face reduction result is that face reduction is needed, determining a target face numerical value corresponding to each sub-model in the three-dimensional model data according to the reference face reduction value;
according to all target surface values corresponding to the sub-models, corresponding surface reduction is carried out on the three-dimensional model data;
compressing the three-dimensional model data subjected to surface reduction into a surface-reduced three-dimensional file and outputting the surface-reduced three-dimensional file;
the subtracting the three-dimensional model data to obtain a subtracting result and a reference subtracting value corresponding to the subtracting result includes:
carrying out plane data collection on the three-dimensional model data to obtain a plurality of plane data; the plane data comprises a plane image, a plane number and a plane size;
inputting all the plane images into a minus plane recognition model, extracting texture features of all the plane images through the minus plane recognition model, and acquiring the minus plane result recognized by the minus plane recognition model according to the extracted texture features;
when the face reduction result is that face reduction is needed, vector conversion is carried out on the number of the plane faces and the plane size in each plane data to obtain a vector matrix corresponding to the three-dimensional model data;
and inputting the vector matrix into a reference recognition model, and performing reference reduced surface prediction on the vector matrix through the reference recognition model to obtain the reference reduced surface value corresponding to the reduced surface result.
2. The method for face reduction of a three-dimensional model according to claim 1, wherein the face reduction recognition of the three-dimensional model to obtain a face reduction result and a reference face reduction value corresponding to the face reduction result further comprises:
carrying out surface number statistics on the three-dimensional model data to obtain the total surface number of the three-dimensional model data;
judging whether the total number of faces is larger than a preset number of faces threshold value;
if the total number of the surfaces is larger than the preset number of the surfaces threshold, determining the surface reduction result as the surface to be reduced;
acquiring a volume value, a surface area value and a size value of each sub-model in the three-dimensional model data;
sequencing the volume values of all the submodels to obtain a first sequencing result, sequencing the surface area values of all the submodels to obtain a second sequencing result, and sequencing the size values of all the submodels to obtain a third sequencing result;
and determining the reference reduced surface value corresponding to the reduced surface result according to the first sorting result, the second sorting result and the third sorting result.
3. The method for face reduction of a three-dimensional model according to claim 1, wherein said inputting all of said planar images into a face reduction recognition model comprises:
acquiring a training sample set; the training sample set comprises a plurality of training samples; each of the training samples is associated with a minus label; the face reduction label comprises a face needing to be reduced and a face needing not to be reduced; one of the training samples comprises six plane sample images in one three-dimensional model sample collected historically;
inputting the training sample into a deep neural network model containing initial parameters;
extracting the texture features of all the plane sample images in the training sample through the deep neural network model;
obtaining a face reduction sample result output by the deep neural network model according to the extracted texture features, and determining a loss value according to the matching degree of the face reduction sample result and the face reduction label;
and when the loss value does not reach the preset convergence condition, iteratively updating the initial parameters of the deep neural network model until the loss value reaches the preset convergence condition, and recording the converged deep neural network model as a trained face-reducing recognition model.
4. The method for face reduction of a three-dimensional model according to claim 1, wherein the plane data collection is performed on the three-dimensional model data to obtain a plurality of plane data; the plane data includes a plane image, a plane number and a plane size, including:
capturing six surfaces of the three-dimensional model data to obtain plane images of the three-dimensional model data in six planes;
summarizing the number of planes of each plane image to obtain the number of planes corresponding to each plane image, and measuring the size of each plane image to obtain the size of each plane corresponding to each plane image;
and determining one plane image, the plane surface number corresponding to the plane image and the plane size as one plane data.
5. The method of reducing a surface of a three-dimensional model of claim 1, wherein determining a target surface value corresponding to each sub-model in the three-dimensional model data from the reference reduced surface value comprises:
obtaining the volume value, the volume ratio, the surface area value, the surface area ratio, the size value and the size ratio of each sub-model in the three-dimensional model data;
determining a protection surface numerical value corresponding to each sub-model according to the volume value, the surface area value and the size value of each sub-model;
multiplying the reference subtraction value by the volume ratio of each submodel to obtain a first ratio corresponding to each submodel, multiplying the reference subtraction value by the surface area ratio of each submodel to obtain a second ratio corresponding to each submodel, and multiplying the reference subtraction value by the size ratio of each submodel to obtain a third ratio corresponding to each submodel;
and determining the target surface value corresponding to the sub-model according to the protection surface value, the first ratio, the second ratio and the third ratio corresponding to the same sub-model.
6. The method for reducing the surface area of the three-dimensional model according to claim 5, wherein the step of determining the target surface value corresponding to the sub-model according to the protection surface value, the first ratio value, the second ratio value and the third ratio value corresponding to the same sub-model comprises the steps of:
rounding the maximum value of the first ratio value, the second ratio value and the third ratio value corresponding to the sub-model, and determining the value as an estimated surface value corresponding to the sub-model;
comparing the estimated surface value corresponding to the sub-model with the protection surface value;
if the estimated surface value corresponding to the sub-model is smaller than or equal to the protection surface value, determining the protection surface value as the target surface value corresponding to the sub-model;
and if the estimated surface value corresponding to the sub-model is larger than the protection surface value, determining the estimated surface value as the target surface value corresponding to the sub-model.
7. A face reducing device for a three-dimensional model is characterized by comprising:
the acquisition module is used for acquiring a three-dimensional file to be processed and constructing three-dimensional model data corresponding to the three-dimensional file to be processed;
the identification module is used for carrying out surface reduction identification on the three-dimensional model data to obtain a surface reduction result and a reference surface reduction value corresponding to the surface reduction result;
the determining module is used for determining a target surface numerical value corresponding to each sub-model in the three-dimensional model data according to the reference surface reduction value when the surface reduction result is the surface reduction required;
the face reduction module is used for carrying out corresponding face reduction on the three-dimensional model data according to all target face numerical values corresponding to the sub models;
the output module is used for compressing the three-dimensional model data subjected to surface reduction into a surface-reduced three-dimensional file and outputting the surface-reduced three-dimensional file;
the identification module is further configured to:
carrying out plane data collection on the three-dimensional model data to obtain a plurality of plane data; the plane data comprises a plane image, a plane number and a plane size;
inputting all the plane images into a minus plane recognition model, extracting texture features of all the plane images through the minus plane recognition model, and acquiring the minus plane result recognized by the minus plane recognition model according to the extracted texture features;
when the face reduction result is that face reduction is needed, vector conversion is carried out on the number of the plane faces and the plane size in each plane data to obtain a vector matrix corresponding to the three-dimensional model data;
and inputting the vector matrix into a reference recognition model, and performing reference reduced surface prediction on the vector matrix through the reference recognition model to obtain the reference reduced surface value corresponding to the reduced surface result.
8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method of de-surfacing a three-dimensional model according to any one of claims 1 to 6 when executing the computer program.
9. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out a method of facelift of a three-dimensional model according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011519504.6A CN112652070B (en) | 2020-12-21 | 2020-12-21 | Three-dimensional model surface reduction method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011519504.6A CN112652070B (en) | 2020-12-21 | 2020-12-21 | Three-dimensional model surface reduction method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112652070A CN112652070A (en) | 2021-04-13 |
| CN112652070B true CN112652070B (en) | 2023-01-13 |
Family
ID=75360278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011519504.6A Active CN112652070B (en) | 2020-12-21 | 2020-12-21 | Three-dimensional model surface reduction method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112652070B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114842122B (en) * | 2022-07-01 | 2022-11-04 | 北京百度网讯科技有限公司 | Model rendering method, device, equipment and storage medium |
| CN117058668B (en) * | 2023-10-10 | 2024-02-02 | 中冶武勘智诚(武汉)工程技术有限公司 | Three-dimensional model face reduction evaluation method and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303597A (en) * | 2015-12-07 | 2016-02-03 | 成都君乾信息技术有限公司 | Patch reduction processing system and processing method used for 3D model |
| CN106504340A (en) * | 2016-11-14 | 2017-03-15 | 成都君乾信息技术有限公司 | A kind of polyhedral model subtracts face method |
| CN106846487A (en) * | 2016-12-20 | 2017-06-13 | 广州爱九游信息技术有限公司 | Subtract face method, equipment and display device |
| CN107895401A (en) * | 2017-11-23 | 2018-04-10 | 塔普翊海（上海）智能科技有限公司 | The data reduction system and its method for simplifying of threedimensional model and application |
| CN108492380A (en) * | 2018-03-29 | 2018-09-04 | 深圳市彬讯科技有限公司 | A kind of threedimensional model subtracts face method and apparatus, storage medium |
| CN109472744A (en) * | 2018-10-29 | 2019-03-15 | 梦工场珠宝企业管理有限公司 | Threedimensional model reduces method |
| CN110287351A (en) * | 2019-06-27 | 2019-09-27 | 哈工大（张家口）工业技术研究院 | A kind of threedimensional model lightweight displaying optimization method |
| CN110363845A (en) * | 2019-07-22 | 2019-10-22 | 南京聚润工程科技有限公司 | A kind of BIM threedimensional model point subtracts the method and system that face stores and reconstructs |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106033621B (en) * | 2015-03-17 | 2018-08-24 | 阿里巴巴集团控股有限公司 | A kind of method and device of three-dimensional modeling |
-
2020
- 2020-12-21 CN CN202011519504.6A patent/CN112652070B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303597A (en) * | 2015-12-07 | 2016-02-03 | 成都君乾信息技术有限公司 | Patch reduction processing system and processing method used for 3D model |
| CN106504340A (en) * | 2016-11-14 | 2017-03-15 | 成都君乾信息技术有限公司 | A kind of polyhedral model subtracts face method |
| CN106846487A (en) * | 2016-12-20 | 2017-06-13 | 广州爱九游信息技术有限公司 | Subtract face method, equipment and display device |
| CN107895401A (en) * | 2017-11-23 | 2018-04-10 | 塔普翊海（上海）智能科技有限公司 | The data reduction system and its method for simplifying of threedimensional model and application |
| CN108492380A (en) * | 2018-03-29 | 2018-09-04 | 深圳市彬讯科技有限公司 | A kind of threedimensional model subtracts face method and apparatus, storage medium |
| CN109472744A (en) * | 2018-10-29 | 2019-03-15 | 梦工场珠宝企业管理有限公司 | Threedimensional model reduces method |
| CN110287351A (en) * | 2019-06-27 | 2019-09-27 | 哈工大（张家口）工业技术研究院 | A kind of threedimensional model lightweight displaying optimization method |
| CN110363845A (en) * | 2019-07-22 | 2019-10-22 | 南京聚润工程科技有限公司 | A kind of BIM threedimensional model point subtracts the method and system that face stores and reconstructs |
Non-Patent Citations (2)
| Title |
|---|
| 基于3DS MAX平台事故再现车辆模型减面优化算法设计;蔡爽等;《重庆理工大学学报(自然科学)》;20200724;第34卷(第11期);第57-62页 * |
| 面向网络化应用的古建筑BIM模型轻量化处理技术;余芳强等;《建筑施工》;20180325;第40卷(第3期);第321-323页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112652070A (en) | 2021-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112652070B (en) | Three-dimensional model surface reduction method, device, equipment and medium | |
| CN110147710B (en) | Method and device for processing human face features and storage medium | |
| CN109711416B (en) | Target identification method and device, computer equipment and storage medium | |
| CN111523414A (en) | Face recognition method and device, computer equipment and storage medium | |
| CN111738351A (en) | Model training method and device, storage medium and electronic equipment | |
| CN116992740B (en) | Intelligent regulation and control method and system for communication base station precise component production equipment | |
| CN111127631A (en) | Single image-based three-dimensional shape and texture reconstruction method, system and storage medium | |
| CN111915572A (en) | Self-adaptive gear pitting quantitative detection system and method based on deep learning | |
| CN111105017A (en) | Neural network quantization method and device and electronic equipment | |
| CN109146000A (en) | A kind of method and device for improving convolutional neural networks based on frost weight | |
| CN111488810A (en) | Face recognition method and device, terminal equipment and computer readable medium | |
| CN113947136A (en) | Image compression and classification method and device and electronic equipment | |
| CN115984662B (en) | Multi-mode data pre-training and identifying method, device, equipment and medium | |
| CN112989910A (en) | Power target detection method and device, computer equipment and storage medium | |
| CN111476269A (en) | Method, device, equipment and medium for constructing balanced sample set and identifying copied image | |
| CN112150612A (en) | Three-dimensional model construction method and device, computer equipment and storage medium | |
| CN112907576A (en) | Vehicle damage grade detection method and device, computer equipment and storage medium | |
| CN113822825B (en) | Optical building target three-dimensional reconstruction method based on 3D-R2N2 | |
| CN108052918A (en) | A kind of person's handwriting Compare System and method | |
| CN117040541B (en) | Data storage method and system for CAD drawing data | |
| CN117095247B (en) | Numerical control machining-based machining gesture operation optimization method, system and medium | |
| CN114492799A (en) | Convolutional neural network model pruning method and device, electronic equipment and storage medium | |
| CN117830321A (en) | Grain quality detection method based on image recognition | |
| CN116403064B (en) | Picture processing method, system, equipment and medium | |
| CN111898493B (en) | Object identification method based on binary quantitative three-dimensional feature descriptor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |