CN102509345B - Portrait art shadow effect generating method based on artist knowledge - Google Patents
Portrait art shadow effect generating method based on artist knowledge Download PDFInfo
- Publication number
- CN102509345B CN102509345B CN201110299880.3A CN201110299880A CN102509345B CN 102509345 B CN102509345 B CN 102509345B CN 201110299880 A CN201110299880 A CN 201110299880A CN 102509345 B CN102509345 B CN 102509345B
- Authority
- CN
- China
- Prior art keywords
- shadow
- portrait
- art
- effect
- artistic
- 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
Landscapes
- Processing Or Creating Images (AREA)
- Image Processing (AREA)
Abstract
The invention provides a portrait art shadow effect generating method based on artist knowledge. By using the invention, the user can conveniently generate the portrait art shadow effect. Firstly, a portrait art shadow templete library is established according to the artist experience. The portrait art shadow effect generating method is implemented based on the templete library. The method comprises the functions of generating art shadow effect according to the art portrait in the library, generating art shadow effect according to the user-defined art portrait, and artifying portrait shadow by one key. The invention can be widely popularized to the fields of interactive digital recreation, movie and television programming, art design and creation and the like.
Description
Technical field
The invention belongs to computing machine shooting field, the particularly method of image processing.Portrait art shadow effect generating method based on artist knowledge specifically.
Background technology
The only Main Means of artist's creative art works, also be the key factor that affects visual art works (photo or drawing etc.) aesthetic feeling, but grasp the skill of using up and need the long-term exercise of artist and training, commonness photograph person or drawing fan are often difficult to shoot or draw out the portrait images with aesthetic, therefore need a kind of method and system of Generative Art effect of shadow in common portrait easily.The object of the invention is to utilize computer technology, make user can generate easily the art shadow effect of portrait.
The work generating about portrait images art shadow effect at present mainly concentrates on three aspects: the shadow effect generating method based on quotient graph or deformable template, the shadow effect generating method based on single image, artistic shadow analysis and evaluation methods.
Aspect the effect of shadow generation based on quotient graph, the Peers of American South University of California etc. utilizes quotient graph to carry out illumination migration to face, static with reference to face object mirror field under different illumination conditions by gathering, and utilize ratio between the facial image of same target under different illumination conditions and the facial image under the uniform illumination condition of front, build the material properties figure of this object under corresponding illumination condition, also referred to as quotient graph.Corresponding expectation illumination condition quotient graph is out of shape simultaneously, and moves on target face, to generate the lighting effect of target image/video scene.The method is for carrying out heavy photo-irradiation treatment of later stage to face video/image.The method limitation is: target face and database face have close geometrical property and material properties, be difficult to be applicable to only have the situation of a width reference picture, because its effect of shadow derives from the pointolite array environment of building in laboratory, be therefore difficult to simulate the effect of shadow of large area light source in artistic photography or art drawing.
Aspect the effect of shadow generation based on deformable template, the Yang wang of Carnegie Mellon Univ USA etc. carry out face image relighting under unknown face geological information and albedo information state.The method input only needs a facial image under any harsh photoenvironment.Utilize a kind of deformable faceform to estimate face geological information, utilized spheric harmonic function base to represent illumination condition, and come the statistical distribution in analog submodule region and the Space Consistency of face texture by Markov random field model.The method is by estimate the illumination condition of facial image with a changeability faceform, and can process complex illumination condition or facial image overexposure and under-exposed situation, can improve greatly the accuracy rate of recognition of face.But the method hypothesis face surface is bright wave surface, and carry out matching facial image with a deformable faceform and estimate geological information, therefore the heavy illumination face image generating lacks certain sense of reality compared with true picture, and is difficult to be applicable to the situation with reference to another piece image effect of shadow.
Aspect the face effect of shadow generation based on single image, the researchist in recognition of face field extracts various illumination invariant feature, for example total variational model, from quotient graph etc., to reach the normalized effect of human face light.These class methods are only applicable to the facial image of low resolution conventionally, and object is to remove face effect of shadow to improve recognition of face rate, rather than generates face effect of shadow true to nature.Intrinsic image technology can by image by pixel resolve into illumination part and reflecting part, but this is an ill-conditioning problem, conventionally automatic computing method are difficult to adapt to the natural image of various complexity, although there is researchist to propose the intrinsic image decomposition method based on user interactions, but be still difficult to the decomposition effect that reaches desirable, thereby the face effect of shadow that cannot directly apply to based on single image generates.The technology of shadow removal aspect also concentrates on the removal of cast shadow at present, also comprises from shade but art shadow effect not only comprises cast shadow conventionally.
Aspect artistic shadow A+E, conventionally comprise based on model (3D shape in image scene is known) and model-free (the 3D shape the unknown in image scene) two class methods, but these two class methods all mainly concentrate on direction and the position of analyzing main light source, and in artistic photography and art drawing, often use the light source of Various Complex shape simultaneously, so current technology is difficult to the art shadow effect of Analysis of Complex, is more difficult to directly apply to art shadow effect and generates.
Summary of the invention
In order to overcome the deficiencies in the prior art, the present invention proposes a kind of portrait art shadow effect generating method based on artist knowledge, can make user generate easily the art shadow effect of portrait; It is characterized in that the method takes following steps:
(1) build portrait art shadow template base based on artist's experience;
(2) portrait art shadow effect based on shadow template base generates.
Further, the described structure portrait art shadow template base of step (1) is taked following steps:
(1.1) be chosen at the representative artistic portrait in illumination aspect according to artist's experience;
(1.2) extract the artistic illumination template of artistic portrait according to artist's experience, i.e. optical mode plate;
(1.3) extract the artistic shade template of artistic portrait according to artist's experience, i.e. shadow template.
Further, the described portrait art shadow effect of step (2) generates the following mode that comprises:
(2.1) artistic portrait Generative Art effect of shadow in reference library;
(2.2) with reference to User Defined art portrait Generative Art effect of shadow;
(2.3) portrait shadow one key is art up.
Further, in the reference library described in mode (2.1), artistic portrait Generative Art effect of shadow is taked following steps:
(2.1.1) utilize moving shape model on face, to locate 160 unique points;
(2.1.2) utilize 160 unique points and multilayer free form morphing, realize artistic shadow template deformation;
(2.1.3) portrait art shadow effect generates.
Further, the described portrait art shadow effect of step (2.1.3) generates and utilizes artist's experience, and shadow template and optical mode plate are separated to deformation process, and the shadow template after utilization distortion and optical mode plate generate the art shadow effect of input portrait.
Further, the described reference User Defined art portrait Generative Art effect of shadow of mode (2.2) is taked following steps:
(2.2.1) calculate the face shadow descriptor of User Defined with reference to artistic portrait;
(2.2.2) coupling one width has the artistic portrait of reference in the storehouse of similar effect of shadow with reference to artistic portrait to User Defined;
(2.2.3) basis, with User Defined with reference to artistic portrait, is adjusted the effect of shadow of match reference art portrait in storehouse automatically, generates the art shadow effect of input portrait.
Further, the described face shadow descriptor of step (2.2.1) is made up of 48 local shadow contrast characteristics, according to face characteristic point location, face is divided into 16 irregular areas, comprise forehead, nose, eyebrow, lip, eyes, cheek; By each region in left and right, up and down, inside and outside in direction, be divided into the subregion of two, and calculate between two sub regions the average value difference on L passage in CIELAB space, face shadow descriptor definition is the vector of one 48 dimension, has comprised the local shadow contrast in the direction of 3, each region in 16 human face regions.
Further, described coupling one width of step (2.2.2) has with reference to artistic portrait to User Defined in the storehouse of similar effect of shadow with reference to artistic portrait, has adopted poor between two face shadow descriptors of Euclidean distance calculating.
Further, the effect of shadow of match reference art portrait in automatic adjustment storehouse described in step (2.2.3), take a kind of automatic adjustment algorithm of shadow template based on face shadow descriptor, the overall shadow contrast of definition portrait is shadow descriptor 48 dimensional feature response sums, and overall shadow contrast difference and shadow descriptor distance according to User Defined with reference to portrait and output portrait images, by adjusting the input and output color range of shadow template and optical mode plate, respectively shadow template is adjusted, thereby adjust the shadow contrast of output portrait, make to export portrait and User Defined approximate with reference to the effect of shadow of portrait.
Further, the described art up following steps of taking of portrait shadow one key of mode (2.3):
(2.3.1) calculate the Shape context distance of inputting each width art portrait in portrait and storehouse;
(2.3.2) coupling one width and input portrait shape of face and face distribute in immediate storehouse with reference to artistic portrait;
(2.3.3) portrait art shadow effect generates.
Compared with prior art, its useful feature is in the present invention: 1,, without calculating input portrait or the three-dimensional geometric information with reference to portrait, only need portrait image to get final product Generative Art effect of shadow; 2, user can freely select the generating mode of three kinds of art shadow effects: artistic portrait Generative Art effect of shadow in reference library, art up with reference to User Defined art portrait Generative Art effect of shadow, portrait shadow one key.
Accompanying drawing explanation:
Fig. 1 is wherein a part of portrait art shadow template schematic diagram of the present invention;
Fig. 2 is artistic portrait Generative Art effect of shadow process schematic diagram in reference library of the present invention;
Fig. 3 is inventor's face characteristic point location schematic diagram
Fig. 4 is that schematic diagram is divided in face of the present invention region;
Fig. 5 is that the present invention is with reference to User Defined art portrait Generative Art effect of shadow process schematic diagram;
Fig. 6 is shadow matching process schematic diagram of the present invention;
Fig. 7 is that shadow template of the present invention is adjusted schematic diagram;
Fig. 8 is the art up process schematic diagram of portrait shadow one key of the present invention;
Fig. 9 is shape of the present invention and face distribution matching process schematic diagram.
Embodiment:
Below in conjunction with the drawings and specific embodiments, the present invention is further detailed explanation.
Any pixel value that the present invention is defined on any passage of the upper rgb space of portrait images P under illumination condition ic is: P (ic)=P (U) * T
i(ic)=P (U) * T
s(ic) * T
l(ic)
Wherein P (U) is the image that this portrait produces under approximate uniform illumination condition U, T
i(ic) be portrait shadow template, it can be decomposed into shadow template T
s(ic) with optical mode plate T
l(ic), shadow template and optical mode plate instruct respectively the pixel on P (U) how to subtract dark and blast.
The present invention utilizes the illumination relevant portion of shadow template approximate representation portrait images, artist utilizes common computer mapping software on a width average face image, to draw optical mode plate and the shadow template with reference to artistic portrait, 500 width art portraits are divided into 10 classes by artist, each class represents a kind of artistic shadow pattern, artist chooses the most representative some drawing portrait shadow templates in each class art portrait, the present invention extracting shadow template from artist's drawing result, build portrait art shadow template base, as shown in Figure 1.
Based on this template base, the present invention has realized portrait art shadow effect generating method and system, and this system is supported the function of three aspects: artistic portrait Generative Art effect of shadow in reference library, art up with reference to User Defined art portrait Generative Art effect of shadow, portrait shadow one key.
Consult Fig. 2, for the artistic portrait of the reference in storehouse, all there is shadow template and optical mode plate corresponding thereto in each width, consult Fig. 3, first the present invention utilizes man face characteristic point positioning method (ASM:Active Shape Model) to locate 160 unique points on input portrait and standard average face, utilizes multilayer free form morphing (MFFD:Multilevel Free-Form Deformation) the shadow template corresponding with the artistic portrait of reference and optical mode plate to be deformed into the shape of input portrait according to the unique point of location:
F
s w＝Warp(F
s，F
d)
According to F
dthe unique point of (d represents target) upper location is by F
s(s represents source) is deformed into F
s w.
For input portrait P, its correspondence is at reference picture P
r(ic) the output image P (ic) under illumination condition ic is:
Wherein
with
for shadow template T corresponding to the artistic portrait of reference
sand optical mode plate T (ic)
l(ic) according to the image after input portrait P distortion.
Consult Fig. 5, if lack the artistic portrait of reference that user specifies in storehouse, system can Auto-matching one width has with reference to artistic portrait to User Defined in the storehouse of similar effect of shadow with reference to artistic portrait, and generates effect of shadow by its corresponding shadow template being carried out automatically adjust.
Have with reference to artistic portrait to User Defined in order to mate a width in the storehouse of similar effect of shadow with reference to artistic portrait, the present invention has designed a kind of face shadow descriptor (FID:Face Illumination Descriptor), distributes to describe face shadow.This face shadow descriptor is made up of 48 local shadow contrast characteristics, consults Fig. 4, and the present invention, according to human face characteristic point, is divided into 16 region { R by face
1, R
2... R
16, comprise the parts such as forehead, nose, eyebrow, lip.The present invention by each region at left and right lr or upper and lower tb or be inside and outside divided into the subregion R of two in cp direction
i1and R
i2, the boundary line of scoring is p
d, and calculate between two sub regions the average value difference on L passage in CIELAB space:
Wherein μ
i1(p
d) and μ
i2(p
d) be respectively two sub regions R
i1and R
i2l passage mean value, d ∈ lr, tb, cp},
for making two sub regions contrast maximum separatrix, r on certain passage mean value in contrast direction
i dfor this contrast characteristic's response.The face shadow descriptor definition of portrait images P is the vector of one 48 dimension, has comprised the local shadow contrast in 3 directions in 16 human face regions:
It is two vectorial Euclidean distances that the present invention defines two differences between face shadow descriptor:
The present invention supposes that the face under identical illumination condition has similar effect of shadow, is in the storehouse of mating with reference to portrait with reference to portrait in the storehouse of the shadow descriptor difference minimum of portrait with User Defined.The process of definition shadow coupling is FIM (P, Ω), in portrait set omega, finds to portrait P and has the portrait of similar effect of shadow, and matching process as shown in Figure 6.
Coupling still there are differences with the artistic portrait of user-defined reference to a certain extent with reference to the effect of shadow of portrait in storehouse, the shadow pattern that has comprised 10 quasi-representatives in the portrait art shadow storehouse that the present invention builds, according to artist's experience, the maximum difference in same shadow pattern is the difference of shadow contrast.The present invention has taked a kind of automatic adjustment algorithm of shadow template based on face shadow descriptor, consults Fig. 7, take shadow template as example, adjusts the input levels maximal value h of shadow template
s(Highest Output Level) and output levels l
s(Lowest Input Level) will directly have influence on the brightness of shadow template, thereby changes the shadow contrast that generates effect.The adjustment of optical mode plate and shadow template class seemingly, utilize the overall contrast of face shadow descriptor FID definition portrait to be:
Might as well establish the artistic portrait of the reference of mating from storehouse is P
m, User Defined is P with reference to artistic portrait
c, input portrait P
ioutput image be P
rif, P
roverall contrast be greater than P
coverall contrast, system will dim optical mode plate and lighten shadow template; If P
roverall contrast be less than P
coverall contrast, system will lighten optical mode plate and dim shadow template.Establish h
sand l
s, h
land l
lrespectively the highest output levels and the minimum input levels of shadow template and optical mode plate.If CT is (FID (P
r)) be greater than CT (FID (P
c)), reduce h
land increase l
s; If CT is (FID (P
r)) be less than CT (FID (P
c)) increase l
lor reduce h
s.
According to artist's experience, the adjustment of color range can be controlled in 20, so, if CT is (FID (P
r)) > CT (FID (P
c)), h
lsetting range be [235,255), l
ssetting range be (0,20].At h
land l
sall 400 in combination in, select and can make P
rand P
cthe net result that face shadow descriptor is adjusted as shadow template apart from a pair of combination of minimum; For contrary situation, if CT is (FID (P
r))≤CT (FID (P
c)), h
ssetting range be [235,255), l
lsetting range be (0,20].At h
sand l
lall 400 in combination in, select and can make P
rand P
cthe net result that face shadow descriptor is adjusted as shadow template apart from a pair of combination of minimum.Utilize the shadow template after adjusting, generate the output image of input portrait, make the effect of shadow of output image and the effect of shadow of User Defined reference picture approximate.
According to artist's experience, different shapes of face is applicable to different effect of shadow, the present invention is based on face characteristic point location (Fig. 3), adopts Shape context distance scale to define shape of face and the face distributional difference of two width portraits.Consult Fig. 8, if user does not have designated reference art portrait, system will be recommended a width and its shape of face and the face artistic portrait of reference (Fig. 9) in approaching storehouse that distributes automatically according to input portrait, and will generate corresponding portrait art shadow effect.
The foregoing is only basic explanations more of the present invention, any equivalent transformation doing according to technical scheme of the present invention, all should belong to protection scope of the present invention.
Claims (5)
1. the portrait art shadow effect generating method based on artist knowledge, can make user generate easily the art shadow effect of portrait; It is characterized in that the method takes following steps:
(1) build portrait art shadow template base based on artist's experience;
(2) portrait art shadow effect based on shadow template base generates;
The described portrait art shadow effect of step (2) generates the following mode that comprises:
(2.1) artistic portrait Generative Art effect of shadow in reference library;
(2.2) with reference to User Defined art portrait Generative Art effect of shadow;
(2.3) portrait shadow one key is art up;
The described reference User Defined art portrait Generative Art effect of shadow of mode (2.2) is taked following steps:
(2.2.1) calculate the face shadow descriptor of User Defined with reference to artistic portrait;
(2.2.2) coupling one width has the artistic portrait of reference in the storehouse of similar effect of shadow with reference to artistic portrait to User Defined;
(2.2.3) basis, with User Defined with reference to artistic portrait, is adjusted the effect of shadow of match reference art portrait in storehouse automatically, generates the art shadow effect of input portrait;
The described face shadow descriptor of step (2.2.1) is made up of 48 local shadow contrast characteristics, according to face characteristic point location, face is divided into 16 irregular areas, comprises forehead, nose, eyebrow, lip, eyes, cheek; By each region in left and right, up and down, inside and outside in direction, be divided into the subregion of two; and calculate between two sub regions the average value difference on L passage in CIELAB space; face shadow descriptor definition is the vector of one 48 dimension, has comprised the local shadow contrast in the direction of 3, each region in 16 human face regions;
In reference library described in mode (2.1), artistic portrait Generative Art effect of shadow is taked following steps:
(2.1.1) utilize moving shape model on face, to locate 160 unique points;
(2.1.2) utilize 160 unique points and multilayer free form morphing, realize artistic shadow template deformation;
(2.1.3) portrait art shadow effect generates;
The art up following steps of taking of portrait shadow one key that mode (2.3) is described:
(2.3.1) calculate the Shape context distance of inputting each width art portrait in portrait and storehouse;
(2.3.2) coupling one width and input portrait shape of face and face distribute in immediate storehouse with reference to artistic portrait;
(2.3.3) portrait art shadow effect generates.
2. the portrait art shadow effect generating method based on artist knowledge as claimed in claim 1, is characterized in that: the described structure portrait art shadow template base of step (1) is taked following steps:
(1.1) be chosen at the representative artistic portrait in illumination aspect according to artist's experience;
(1.2) extract the artistic illumination template of artistic portrait according to artist's experience, i.e. optical mode plate;
(1.3) extract the artistic shade template of artistic portrait according to artist's experience, i.e. shadow template.
3. the portrait art shadow effect generating method based on artist knowledge as claimed in claim 3, it is characterized in that: the described portrait art shadow effect of step (2.1.3) generates the artist's experience of utilizing, shadow template and optical mode plate are separated to deformation process, utilize shadow template and optical mode plate after distortion to generate the art shadow effect of inputting portrait.
4. the portrait art shadow effect generating method based on artist knowledge as claimed in claim 1, it is characterized in that: described coupling one width of step (2.2.2) has with reference to artistic portrait to User Defined in the storehouse of similar effect of shadow with reference to artistic portrait, adopted poor between two face shadow descriptors of Euclidean distance calculating.
5. the portrait art shadow effect generating method based on artist knowledge as claimed in claim 1, it is characterized in that: the effect of shadow of match reference art portrait in the automatic adjustment storehouse described in step (2.2.3), take a kind of automatic adjustment algorithm of shadow template based on face shadow descriptor, the overall shadow contrast of definition portrait is shadow descriptor 48 dimensional feature response sums, and overall shadow contrast difference and shadow descriptor distance according to User Defined with reference to portrait and output portrait images, by adjusting the input and output color range of shadow template and optical mode plate, respectively shadow template is adjusted, thereby adjust the shadow contrast of output portrait, make to export portrait and User Defined approximate with reference to the effect of shadow of portrait.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110299880.3A CN102509345B (en) | 2011-09-30 | 2011-09-30 | Portrait art shadow effect generating method based on artist knowledge |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110299880.3A CN102509345B (en) | 2011-09-30 | 2011-09-30 | Portrait art shadow effect generating method based on artist knowledge |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102509345A CN102509345A (en) | 2012-06-20 |
| CN102509345B true CN102509345B (en) | 2014-06-25 |
Family
ID=46221422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110299880.3A Active CN102509345B (en) | 2011-09-30 | 2011-09-30 | Portrait art shadow effect generating method based on artist knowledge |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102509345B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104639843B (en) * | 2014-12-31 | 2017-12-05 | 小米科技有限责任公司 | Image processing method and device |
| CN111063017B (en) * | 2018-10-15 | 2022-04-12 | 华为技术有限公司 | Illumination estimation method and device |
| CN109300170B (en) * | 2018-10-18 | 2022-10-28 | 云南大学 | Method for transmitting shadow of portrait photo |
| CN111382618B (en) | 2018-12-28 | 2021-02-05 | 广州市百果园信息技术有限公司 | Illumination detection method, device, equipment and storage medium for face image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034481A (en) * | 2007-04-06 | 2007-09-12 | 湖北莲花山计算机视觉和信息科学研究院 | Method for automatically generating portrait painting |
| CN101950352A (en) * | 2010-05-31 | 2011-01-19 | 北京智安邦科技有限公司 | Target detection method capable of removing illumination influence and device thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090046099A1 (en) * | 2006-11-13 | 2009-02-19 | Bunkspeed | Real-time display system |
| KR100897385B1 (en) * | 2007-06-27 | 2009-05-14 | 성균관대학교산학협력단 | Method and apparatus for illumination normalization |
-
2011
- 2011-09-30 CN CN201110299880.3A patent/CN102509345B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034481A (en) * | 2007-04-06 | 2007-09-12 | 湖北莲花山计算机视觉和信息科学研究院 | Method for automatically generating portrait painting |
| CN101950352A (en) * | 2010-05-31 | 2011-01-19 | 北京智安邦科技有限公司 | Target detection method capable of removing illumination influence and device thereof |
Non-Patent Citations (4)
| Title |
|---|
| 《Learning Artistic Lighting Template from Portrait Photographs》;Xin Jin et al.;《ECCV 2010, PART IV, LNCS 6314》;20101231;101-114 * |
| 《超酷光影效果轻松造》;青山漫步;《电脑迷》;20071001(第79期);63 * |
| Xin Jin et al..《Learning Artistic Lighting Template from Portrait Photographs》.《ECCV 2010, PART IV, LNCS 6314》.2010,101-114. |
| 青山漫步.《超酷光影效果轻松造》.《电脑迷》.2007,(第79期),63. |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102509345A (en) | 2012-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhou et al. | Sparsefusion: Distilling view-conditioned diffusion for 3d reconstruction | |
| CN106780293B (en) | Style transfer of head close-up portrait | |
| Neubert et al. | Approximate image-based tree-modeling using particle flows | |
| CN106067190B (en) | A kind of generation of fast face threedimensional model and transform method based on single image | |
| CN106920274A (en) | Mobile terminal 2D key points rapid translating is the human face model building of 3D fusion deformations | |
| CN109816784B (en) | Method and system for three-dimensional reconstruction of human body and medium | |
| Wu et al. | Making bas-reliefs from photographs of human faces | |
| CN107067429A (en) | Video editing system and method that face three-dimensional reconstruction and face based on deep learning are replaced | |
| Ji et al. | Bas-relief modeling from normal images with intuitive styles | |
| CN102509345B (en) | Portrait art shadow effect generating method based on artist knowledge | |
| CN109325994B (en) | Method for enhancing data based on three-dimensional face | |
| CN112258387A (en) | Image conversion system and method for generating cartoon portrait based on face photo | |
| CN103262126A (en) | Image processor, lighting processor and method therefor | |
| CN111626951A (en) | Image shadow elimination method based on content perception information | |
| Malleson et al. | Rapid one-shot acquisition of dynamic VR avatars | |
| Faustino et al. | Simple adaptive mosaic effects | |
| Zhang et al. | Portrait relief modeling from a single image | |
| CN110363804B (en) | Flower bas-relief generating method based on deformation model | |
| CN113436058B (en) | Character virtual clothes changing method, terminal equipment and storage medium | |
| CN117157673A (en) | Method and system for forming personalized 3D head and face models | |
| CN113065506B (en) | Human body posture recognition method and system | |
| Lee et al. | High-quality depth estimation using an exemplar 3d model for stereo conversion | |
| CN115661367B (en) | Dynamic hybrid deformation modeling method and system based on photo collection | |
| Lin et al. | Video stylization: painterly rendering and optimization with content extraction | |
| Liang et al. | A new face relighting method based on edge-preserving filter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |