CN115082597A - Palette-based image recoloring method and system - Google Patents
Palette-based image recoloring method and system Download PDFInfo
- Publication number
- CN115082597A CN115082597A CN202210468112.4A CN202210468112A CN115082597A CN 115082597 A CN115082597 A CN 115082597A CN 202210468112 A CN202210468112 A CN 202210468112A CN 115082597 A CN115082597 A CN 115082597A
- Authority
- CN
- China
- Prior art keywords
- palette
- color
- convex hull
- image
- recoloring
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/40—Filling a planar surface by adding surface attributes, e.g. colour or texture
Abstract
The invention provides a palette-based image recoloring method and system, and relates to the technical field of image processing. The proposed method comprises: extracting an RGB space convex hull corresponding to an input image; clustering in the RGB space convex hull to obtain a clustering center of image pixel points, screening the clustering center as a representative color, and forming a color palette by the convex hull vertex and the representative color corresponding to the clustering center; performing tetrahedral subdivision on a vertex set corresponding to the color palette, and calculating interpolation weight of the pixel points about the color of the color palette; and modifying the color of the color palette, and realizing image recoloring editing by utilizing the interpolation weight. The extracted palette has better representativeness and editing locality, and is convenient for a user to locally edit the image.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a system for recoloring an image based on a palette.
Background
The heavy coloring is a popular research direction in the field of visual media editing, is widely applied to movie production and art design, and plays an important role in color editing of image and video.
At present, two types of image recoloring methods based on color palette are available: the image palettes are extracted by clustering and by computing convex hulls. Among them, extracting an image palette by calculating a convex hull is the latest technology in this field. At present, the method of extracting the palette of an image by calculating a convex hull is to regard the image as a point set of an RGB color space, calculate the convex hull of the point set and further simplify the calculation. Then, the weight of each pixel point about the color palette is obtained through an MVC interpolation method, and finally, the image re-coloring is realized by modifying the colors of the color palette, so that the hidden representative colors can be extracted, but some representative colors in the image may be located in the convex hull and ignored, that is, the representative colors located in the convex hull are ignored by the existing color palette-based image re-coloring method, so that the color palette is lack of representativeness on the whole and is not beneficial to subsequent image editing.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a palette-based image recoloring method and a palette-based image recoloring system, which solve the technical problem that the prior method ignores the representative colors positioned in the convex hull.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme:
(III) advantageous effects
The invention provides a palette-based image recoloring method and system. Compared with the prior art, the method has the following beneficial effects:
the invention relates to a method and a system for recoloring an image based on a palette, wherein the method comprises the steps of firstly extracting RGB space convex hulls corresponding to an input image; then clustering in the RGB space convex hull to obtain a clustering center of image pixel points, screening the clustering center as a representative color, and forming a color palette by the convex hull peak and the representative color corresponding to the clustering center; performing tetrahedral subdivision on a vertex set corresponding to the color palette, and calculating interpolation weight of the pixel points about the color of the color palette; and finally, modifying the color of the color palette, and realizing image recoloring editing by utilizing the interpolation weight. The extracted palette has better representativeness and editing locality, and is convenient for a user to locally edit the image.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a block flow diagram of an embodiment of the present invention;
FIG. 2 is a diagram of an algorithm framework according to an embodiment of the present invention;
FIG. 3 is a simplified diagram of a convex hull according to an embodiment of the present invention;
fig. 4a and 4b are both effect graphs of the verification experiment, in which the 1 st column is input, the 2 nd and 3 rd columns are effect graphs of the existing algorithm, and the 4 th column is an effect graph of the method according to the embodiment of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer and more complete description of the technical solutions in the embodiments of the present invention, it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the application provides the image recoloring method and the image recoloring system based on the color palette, so that the technical problem that the representative color in the convex hull is ignored in the existing method is solved, the extracted color palette has better representativeness and editing locality, and the recoloring effect is superior.
In order to solve the technical problems, the general idea of the embodiment of the application is as follows:
at present, two types of image recoloring methods based on color palette are available: the first method extracts the image palette through clustering, then obtains the weight of each pixel point about a clustering center through RBF interpolation to perform image recolouring editing, and the representative color extracted by the method completely comes from the image, thereby well representing the color distribution of the image. However, representative colors in an image do not necessarily appear explicitly on the image. The second method extracts the image palette by calculating a convex hull. Such methods treat the image as a set of points in the RGB color space, whose convex hull is computed and further simplified. Then, the weight of each pixel point relative to the color palette is obtained through an MVC interpolation method, and finally, the image recoloring is realized by modifying the colors of the color palette. In order to solve the above problem, in the embodiment of the present invention, an internal representative vertex (color) is added on the basis of a simplified convex hull of an image to enhance the expression capability of a color palette, and meanwhile, the colors (vertices) of the color palette are spatially subdivided, and pixels are interpolated in each tetrahedral to improve sparsity of interpolation weights, so that the extracted color palette has better representativeness, the interpolation weights are sparser, and more accurate local editing is supported.
In order to better understand the technical solution, the technical solution will be described in detail with reference to the drawings and the specific embodiments.
An embodiment of the present invention provides a method for re-coloring an image based on a palette, as shown in fig. 1 to 2, including:
s1, extracting RGB space convex hulls corresponding to the input images;
s2, clustering in the RGB space convex hull to obtain a clustering center of image pixel points, screening the clustering center as a representative color, and enabling the convex hull vertex and the representative color corresponding to the clustering center to jointly form a color palette;
s3, performing tetrahedral subdivision on the vertex set corresponding to the color palette, and calculating the interpolation weight of the pixel points about the color of the color palette;
and S4, modifying the color of the color palette, and realizing image recoloring editing by using the interpolation weight.
The individual steps are described in detail below:
in step S1, the RGB spatial convex hull corresponding to the input image is extracted. The specific implementation process is as follows:
the input image is regarded as a point set of an RGB space, an original three-dimensional convex hull of the point set is calculated, the original three-dimensional convex hull often comprises hundreds of vertexes, the image cannot be directly edited as a palette, the image needs to be simplified, and the simplification process is as follows:
as shown in fig. 3, the edge (v) is divided i ，v j ) Reduced to one vertex v. In particular, one needs to find a vertex v outside the convex hull, so that v is connected (v;) i ，v j ) The new convex hull formed by the adjacent points of (v) expands outwards and is completely wrapped i ，v j ) Thus, the edge (v) is reduced i ，v j ) And the corresponding two triangular patches are deleted from the original convex hull, so that the topological simplification of the convex hull is realized while the volume of the convex hull is expanded. Wherein the expression of v is as follows:
the right side of the equation represents the volume of the convex hull expansion increase caused by the contraction edge, and A represents (v) i ，v j ) And (2) associating the area of each triangular patch, wherein n represents the normal direction of each triangular patch, v is the edge contraction for enumerating all edges of the convex hull each time in the actual operation of the vertex coordinates to be solved and selecting the edge with the smallest volume increase, and the algorithm is ended when the volume of the convex hull is increased to a certain extent (in the specific implementation process, along with the increase of the volume of the convex hull, the reconstruction error is increased, and when the reconstruction error is larger than 2/255, the expansion of the convex hull is stopped). And obtaining a convex hull with a small number of top points after the algorithm is finished, namely an RGB space convex hull.
In step S2, clustering is performed in the RGB space convex hull to obtain a cluster center of image pixels, and the cluster center is screened as a representative color, and the peaks of the convex hull and the representative colors corresponding to the cluster center form a color palette. The specific implementation process is as follows:
in step S1, as the edge reduction continues, the volume of the convex hull increases, and the representative colors inside the convex hull cannot be added to the color palette and cannot participate in the subsequent image editing task. In order to make the palette more representative, the embodiment of the present invention further optimizes the initial palette by clustering the representative colors that are ignored and located inside the convex hull.
The embodiment of the invention clusters the image pixel points by using a MeanShift algorithm to extract the representative color in the image. It should be noted that other clustering such as hierarchical clustering can also be performed, and the embodiment of the present invention introduces the use of the MeanShift algorithm to perform clustering on image pixels in detail.
Regarding each pixel point in the RGB space convex hull as an independent clustering center, then carrying out mean shift on each pixel point, finally converging each pixel point to a fixed position through multiple rounds of mean shift, and finally classifying the pixel points with close positions into the same class. Specifically, for an arbitrary pixel point, the coordinate in the RGB space is assumed to be c 0 ＝p 0 (R 0 ，G 0 ，B 0 ) Using r as radius to form a sphere S 0 Then calculate the falling at S 0 Mean value c of all pixels in 1 ＝(R 1 ，G 1 ，B 1 ) Then again with c 1 Making the sphere S with the sphere center r as the radius 1 And calculating the average value of the pixel points in the sphere again, and continuously repeating the process until S kt1 Center of sphere c k+1 Distance S k Center of sphere c k Close enough together the algorithm is considered to converge and stop the mean shift. The formalized representation is as follows:
wherein p is 1 ，p 2 ，…，p M Is located at the center of the sphere as c k All the pixel points in the sphere with the radius r.
The estimation of the radius r is particularly important for MeanShift clustering, and generally, the radius is small, so that more types are obtained finally, the radius is large, and the number of types obtained finally is small. The radius selection method in the embodiment of the invention comprises the following steps:
firstly, randomly selecting a plurality of pixel points in an image, calculating the distances of all the point pairs and sequencing the point pairs, wherein the sequenced sequence is marked as D. Assuming that the number of all the point pairs is N, the distance r at position Nq in D is taken as the radius of the mean shift.
The efficiency of directly performing MeanShift clustering on all pixels is very low, and in order to increase the speed, the embodiment of the invention samples the original image, and then only considers the points on the sampling point set as the initial clustering center to perform mean shift. After clustering, some new representative colors are obtained, which are basically located inside the convex hull of the RGB space obtained in S1 in the RGB space, and some colors may be very close to the vertex of the convex hull, so that the color palette is redundant. To this end, the colors of all the initial palettes are retained and those colors in the clustering result that are similar to the initial palettes are deleted. Specifically, the algorithm deletes those colors that are less than r away from the convex hull vertex, resulting in a representative color.
And the convex hull vertex and the representative colors corresponding to the clustering centers form a color palette together.
In step S3, a vertex set corresponding to the palette is tetrahedrally subdivided, and interpolation weights of the pixel points with respect to the colors of the palette are calculated, where the vertex set includes a convex hull vertex and a cluster center point. The specific implementation process is as follows:
and organizing the clustering center points corresponding to all the representative colors and the convex hull vertexes together to form a three-dimensional vertex set, and performing Delaunay tetrahedron subdivision on the three-dimensional vertex set. Thus, the set of image pixel points will be naturally segmented into the interior of the plurality of tetrahedrons.
For interpolation weight
At present, the calculation is mainly divided into optimization-based and interpolation-based methods. Interpolation weights obtained by an optimization-based method have good smoothness, but the calculation efficiency is low, tens of minutes are required for calculating a single image, and a large number of control points are required to be provided by the current interpolation-based method. Therefore, when adjustingWhen the number of color plates is small, sparse interpolation is difficult to realize by the existing algorithm, that is, interpolation weights of pixel points about most color palettes are all larger than 0, and modifying some color palettes may cause large-area color change of an image, so that effective local editing is difficult to realize.
According to the embodiment of the invention, after the tetrahedron is divided, each pixel is only related to 4 vertexes of the tetrahedron, so that the interpolation weight of each point is only 4 elements with the value larger than 0 at most, and the sparsity requirement is naturally met, thereby realizing better local editing.
After the color vertex space of the color palette is subdivided, a group of spatial tetrahedrons is obtained, and interpolation aiming at the tetrahedrons has good definition. Specifically, in a certain tetrahedron ABCD after the color vertex space of the palette is subdivided, the interpolation weight of the vertex a corresponding to any pixel point p can be simply expressed as:
where T (pBCD) represents the volume of a tetrahedron composed of 4 vertices of pBCD, and T (ABCD) represents the volume of a tetrahedron ABCD, and similarly, interpolation weights for p points with respect to the other 3 vertices of the tetrahedron can be obtained
And
in step S4, the colors of the palette are modified, and image recoloring editing is implemented using the interpolation weights. The specific implementation process is as follows:
for convenience of description, a color set of the palette is denoted as V and a pixel point set of the input image is denoted as P. In order to realize image editing, interpolation is carried out on all pixel points of an input image. Such as formula
And expressing any pixel point P epsilon P as a convex combination of the palette V. Wherein the content of the first and second substances,
representing pixel p with respect to palette color V i And satisfy the interpolation weight of
In the actual image editing process, the interpolation weight is calculated only once at the beginning and then is kept unchanged, and the user only needs to modify one or more colors of the color palette V and substitutes the modified color palette V' into the formula
The edited image color can be obtained.
To verify the validity of the embodiments of the present invention, the following verification was performed:
the experimental environment is as follows: win10, AMDR7-5800H, 16GB memory. In order to verify the sensitivity of the parameters, firstly, the parameter q used by the MeanShift algorithm is verified; then, in order to illustrate the excellent sparsity of the interpolation weight of the algorithm of the embodiment of the invention, the method of the embodiment of the invention is compared with the prior K-Means-based image recoloring algorithm; next, to illustrate that the algorithm of the embodiment of the present invention supports better local editing, the image recoloring effect of the method of the embodiment of the present invention is compared with that of the existing image recoloring algorithm based on convex hull optimization. The experimental results show that the color palette extracted by the algorithm of the embodiment of the invention has better representativeness, compared with the current best algorithm, the interpolation weight of the embodiment of the invention is more sparse, the image local editing is more facilitated, and the better recoloring effect is realized. As shown in fig. 4(a), the user's editing intention is that the color of the plant in the upper right corner changes from green to brown. The method shown in the second column changes the color of the flower and the color of the pistil while modifying the color of the plant. The method shown in the third column causes unnatural changes in the color of the ground in the image. The embodiment of the invention well realizes the editing intention and hardly influences other areas of the image. As shown in fig. 4(b), the user's editing intention is to modify the lady's brown hair into blue to further highlight the hair dyeing effect. The existing methods all introduce unnatural facial color changes, but the method of the embodiment of the invention accurately realizes the hair dyeing effect.
The embodiment of the invention also provides an image recoloring system based on the palette, which comprises:
the convex hull obtaining module is used for extracting an RGB space convex hull corresponding to the input image;
the clustering module is used for clustering in the RGB space convex hull to obtain a clustering center of image pixel points, screening the clustering center as a representative color, and forming a color palette by the convex hull vertex and the representative color corresponding to the clustering center;
the subdivision module is used for carrying out tetrahedral subdivision on the vertex set corresponding to the color palette and calculating the interpolation weight of the pixel points about the color of the color palette;
and the recoloring module is used for modifying the color of the color palette and realizing image recoloring editing by utilizing the interpolation weight.
It is to be understood that the palette-based image recoloring system provided in the embodiments of the present invention corresponds to the above palette-based image recoloring method, and for the explanation, examples, and beneficial effects of the related contents, reference may be made to the corresponding contents in the palette-based image recoloring method, which is not described herein again.
Embodiments of the present invention also provide a computer-readable storage medium storing a computer program for palette-based image recoloring, where the computer program causes a computer to execute the palette-based image recoloring method as described above.
An embodiment of the present invention further provides an electronic device, including:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs comprising instructions for performing the palette-based image recoloring method as described above.
In summary, compared with the prior art, the method has the following beneficial effects:
1. the extracted palette has better representativeness and editing locality, and is convenient for a user to locally edit the image.
2. The embodiment of the invention carries out space subdivision on the vertex set of the color palette and interpolates the image pixels by using the barycentric coordinates, thereby obviously improving the sparsity of interpolation weights.
It is to be noted that, in the embodiments of the present invention, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will 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; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for palette-based image recoloring, comprising:
s1, extracting RGB space convex hulls corresponding to the input images;
s2, clustering in the RGB space convex hull to obtain a clustering center of image pixel points, screening the clustering center as a representative color, and enabling the convex hull vertex and the representative color corresponding to the clustering center to jointly form a color palette;
s3, performing tetrahedral subdivision on the vertex set corresponding to the color palette, and calculating the interpolation weight of the pixel points about the color of the color palette;
and S4, modifying the color of the color palette, and realizing image recoloring editing by using the interpolation weight.
2. The method of claim 1, wherein extracting the RGB spatial convex hull corresponding to the input image comprises:
regarding an input image as a point set of an RGB space, and calculating an original three-dimensional convex hull of the point set;
combining the edges (v) of the original three-dimensional convex hull i ，v j ) Reducing the distance to a vertex v to obtain an RGB space convex hull, wherein the calculation formula is as follows:
the right side of the equation represents the volume of the convex hull expansion increase caused by the contraction edge, and A represents (v) i ，v j ) And (2) associating the area of each triangular patch, wherein n represents the normal direction of each triangular patch, v is the contraction of all edges of each enumerated convex hull in the practical operation of the vertex coordinates to be solved and the edge with the minimum volume increase is selected, the reconstruction error is increased along with the increase of the volume of the convex hull, and the algorithm is ended when the reconstruction error exceeds a preset threshold value to obtain the RGB space convex hull.
3. The method of claim 1, wherein clustering in the RGB convex hull to obtain cluster centers for image pixels comprises:
regarding each pixel point in the RGB space convex hull as an independent clustering center, then carrying out mean shift on each pixel point, converging each pixel point to a fixed position through multiple rounds of mean shift, and classifying the pixel points with close positions into the same class.
4. The method of claim 1, wherein said filtering the cluster centers as representative colors comprises:
and deleting the clustering center of which the distance from the convex hull vertex is less than r, wherein r is the radius of mean shift.
5. The method of palette-based image recoloring of claim 1, wherein the method of tetrahedrally subdividing the set of vertices corresponding to the palette comprises:
and organizing the clustering center points of all the representative colors and the convex hull vertexes together to form a vertex set, and performing Delaunay tetrahedron subdivision on the vertex set.
6. The method of claim 1, wherein the calculating interpolation weights for pixel points with respect to palette colors comprises:
inside the tetrahedral ABCD, the interpolation weight of the vertex a corresponding to any pixel point p can be simply expressed as:
wherein T (pBCD) represents the volume of a tetrahedron composed of 4 vertices of pBCD, and T (ABCD) represents the volume of a tetrahedral ABCD.
7. The method of claim 1, wherein modifying the colors of the palette comprises:
marking a color set of the palette as V and a pixel point set of the input image as P;
interpolating all pixel points of the input image, as a formula:
expressing any pixel point P E P as a convex combination of the palette V; wherein the content of the first and second substances,
indicating a pixel p with respect to the palette color V i And satisfy the interpolation weight of
Modifying one or more colors of the palette V and substituting the modified palette V' into the formula
The edited image color can be obtained.
8. A palette-based image recoloring system, the system comprising:
the convex hull obtaining module is used for extracting an RGB space convex hull corresponding to the input image;
the clustering module is used for clustering in the RGB space convex hull to obtain a clustering center of image pixel points, screening the clustering center as a representative color, and forming a color palette by the convex hull vertex and the representative color corresponding to the clustering center;
the subdivision module is used for carrying out tetrahedral subdivision on the vertex set corresponding to the color palette and calculating the interpolation weight of the pixel points about the color of the color palette;
and the recoloring module is used for modifying the color of the color palette and realizing image recoloring editing by utilizing the interpolation weight.
9. A computer-readable storage medium storing a computer program for palette-based image recoloring, wherein the computer program causes a computer to execute the palette-based image recoloring method according to any one of claims 1 to 7.
10. An electronic device, comprising:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured for execution by the one or more processors, the programs comprising instructions for performing the palette-based image recoloring method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210468112.4A CN115082597A (en) | 2022-04-29 | 2022-04-29 | Palette-based image recoloring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210468112.4A CN115082597A (en) | 2022-04-29 | 2022-04-29 | Palette-based image recoloring method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115082597A true CN115082597A (en) | 2022-09-20 |
Family
ID=83247966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210468112.4A Pending CN115082597A (en) | 2022-04-29 | 2022-04-29 | Palette-based image recoloring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115082597A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115690249A (en) * | 2022-11-03 | 2023-02-03 | 武汉纺织大学 | Method for constructing digital color system of textile fabric |
-
2022
- 2022-04-29 CN CN202210468112.4A patent/CN115082597A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115690249A (en) * | 2022-11-03 | 2023-02-03 | 武汉纺织大学 | Method for constructing digital color system of textile fabric |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11257279B2 (en) | Systems and methods for providing non-parametric texture synthesis of arbitrary shape and/or material data in a unified framework | |
| Lozes et al. | Partial difference operators on weighted graphs for image processing on surfaces and point clouds | |
| WO2013106984A1 (en) | Learning painting styles for painterly rendering | |
| CN110570440A (en) | Image automatic segmentation method and device based on deep learning edge detection | |
| US20230306685A1 (en) | Image processing method, model training method, related apparatuses, and program product | |
| Kawasaki et al. | An image processing approach to feature-preserving B-spline surface fairing | |
| CN115661404A (en) | Multi-fine-grain tree real scene parametric modeling method | |
| CN115082597A (en) | Palette-based image recoloring method and system | |
| Gai et al. | Artistic low poly rendering for images | |
| Deng et al. | ELMGAN: A GAN-based efficient lightweight multi-scale-feature-fusion multi-task model | |
| CN110322548B (en) | Three-dimensional grid model generation method based on geometric image parameterization | |
| CN108241861A (en) | A kind of data visualization method and equipment | |
| CN116703744A (en) | Remote sensing image dodging and color homogenizing method and device based on convolutional neural network | |
| Bayazit | A greedy region growing algorithm for anisotropic stretch adaptive triangulation of geometry images | |
| Chaurasia et al. | Editable parametric dense foliage from 3D capture | |
| Xing et al. | Highly parallel algorithms for visual-perception-guided surface remeshing | |
| Beardsley et al. | Editable parametric dense foliage from 3D capture | |
| CN113034560A (en) | Non-uniform texture migration method and device | |
| Zeng et al. | 3D plants reconstruction based on point cloud | |
| Pan et al. | Salient structural elements based texture synthesis | |
| CN117710603B (en) | Unmanned aerial vehicle image three-dimensional building modeling method under constraint of linear geometry | |
| Lo et al. | Generating Color Scribble Images using Multi‐layered Monochromatic Strokes Dithering | |
| Ma-thi et al. | Image-Based Approach for Preparing Artifacts for Virtual Puppetry Museum | |
| Chen et al. | Adaptive Surface Texture Synthesis Using Round‐Shaped Neighbourhoods | |
| Zhang et al. | Simplicity Driven Edge Refinement and Color Reconstruction in Image Vectorization |
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 |