CN101266685A - A method for removing unrelated images based on multiple photos - Google Patents
A method for removing unrelated images based on multiple photos Download PDFInfo
- Publication number
- CN101266685A CN101266685A CNA2007100643896A CN200710064389A CN101266685A CN 101266685 A CN101266685 A CN 101266685A CN A2007100643896 A CNA2007100643896 A CN A2007100643896A CN 200710064389 A CN200710064389 A CN 200710064389A CN 101266685 A CN101266685 A CN 101266685A
- Authority
- CN
- China
- Prior art keywords
- window
- unrelated images
- photo
- background
- images
- 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
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000009466 transformation Effects 0.000 claims abstract description 13
- 239000000463 material Substances 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 230000000295 complement effect Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 7
- 239000000284 extract Substances 0.000 description 5
- 239000004744 fabric Substances 0.000 description 5
- 230000008878 coupling Effects 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 4
- 238000005859 coupling reaction Methods 0.000 description 4
- 238000012876 topography Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000012467 final product Substances 0.000 description 2
- 241000282320 Panthera leo Species 0.000 description 1
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention presents a method of removing irrespective images based on a plurality of photographs, which utilizes complementary information in a plurality of photographs shot by users to remove irrespective images in the photographs, and can restore the real situation of the background shaded by tourists so as to keep the authenticity of the content of the photograph background. The users only need to specify a window containing irrespective images in a photograph, the invention can automatically search alternative windows matched with the window in a plurality of photographs with computer vision technique, and automatically carries out geometric transformation for the alternative windows so as to causes them to be exactly matched with the background around the irrespective image in the original photograph. The users only need to choose an alternative window containing only the background, which is removing the irrespective images from the photograph and completely restoring the real situation of the background shaded by tourists at the same time. The invention can also be used to remove other irrespective moving objects in the photograph, such as automobiles, etc.
Description
Technical field
The invention belongs to technical field of image processing, relate to the method for the removal unrelated images of multiple photos.
Background technology
The scenic spot visitor who is in the tourist season is a lot of usually, often has incoherent visitor in the photo and appears in the background.Want to wait for that all visitors leave needs long time or impossible at all from picture.Visitor in the background together can only be taken when much taking pictures, this has often destroyed the integrality and the aesthetic property of picture.The visitor who removes in the photo background becomes a technology of great practical value.
From an independent photo, remove the visitor, remain a very difficult task at present, because the scene information that is sheltered from by the visitor is to recover fully truly.Along with the progress of digital camera and digital memory technology, take multiple photos in same scene and become economically and can accept.In real life, the mobile of visitor is quite frequent in the sight spot, and a visitor does not have yet in the moment background although be difficult to find, and in the different moment, because the visitor's is mobile, the situation that scenery is blocked by the visitor is constantly changing.At some constantly, certain part scenery is blocked by the visitor, but in the other moment, this part scenery intactly shows.If constantly take multiple photos the different of roughly the same angle for same scene, so, utilize the relation of replenishing mutually between this group photo, as long as certain background parts is not to be blocked at least, so just can recover this part background on a photos.
Summary of the invention
Take pictures in the tourism scene that some visitor is numerous, can't wait until in the moment scene without any the visitor, at this moment can only photograph the photo that comprises the visitor, in order to address this problem, the objective of the invention is to utilize the complementary information in the multiple photos, in a spot of presence of user, unrelated images in the photo is removed, and the truth of the background that can be blocked by unrelated images is recovered out, the authenticity that keeps photo content, for this reason, the present invention proposes a kind of method of the removal unrelated images based on multiple photos.
To achieve these goals, the method that the present invention is based on the removal unrelated images of multiple photos comprises following steps,
Step S1: the user takes multiple photos to a scene;
Step S2: choose and comprise the minimum photo of unrelated images in the multiple photos as main photo;
Step S3: judging has the unrelated images of denying to need to remove in the main photo, need to remove if having, and then execution in step 4; If do not need to remove, then execution in step 8;
Step S4: the selected unrelated images window that will remove unrelated images in main photo;
Step S5: utilize automated graphics registration technology matching image feature, calculate unrelated images window corresponding position in every non-main photo, and the image of these positions is generated a plurality of replacement windows through geometric transformation;
Step S6: in a plurality of replacement windows, select a replacement window that only comprises background;
Step S7: the picture material in the replacement window of selection is revised main photo through the picture material of replacing in the unrelated images window after the geometric transformation, and execution in step S3 revises main photo successively;
Step S8: successively all unrelated images in the main photo are removed one by one, obtained a background pictures that does not comprise unrelated images.
Preferably, described multiple photos take be from a roughly the same background the position, according to the angle of roughly the same background, take at a certain time interval; In such multiple photos, occur once on the photo at least therein of each the part background in the scene.
Preferably, described unrelated images window be shaped as rectangle, only need comprise unrelated images to small part, the size of this unrelated images window is enlarged to comprise the background information around the unrelated images automatically by algorithm.
Preferably, described automated graphics registration uses the SIFT characteristics of image to carry out feature extraction and characteristic matching, and uses the geometric transformation parameter between interative least square method estimation unrelated images window and the replacement window.
The invention has the advantages that:
(1) mobile object that can the unrelated images in the photo or other are irrelevant is removed, and the truth of the background that can be blocked by unrelated images recovers out, keeps the authenticity of photo content.
(2) performance of this method is irrelevant with the complicacy of the background of being blocked by unrelated images, and situation that comprise a lot of detailed shape very big for the zone that is blocked still is suitable for well.
(3) simple and convenient user operation need not accurately to delineate out the unrelated images profile that will remove, only need roughly specify a rectangular window that comprises unrelated images to get final product in image.
Description of drawings:
Fig. 1 is a process flow diagram, has shown the algorithmic procedure of image processing section of the present invention.
Fig. 2 is one group of example image of input algorithm.
Fig. 3 is the intermediate result in the algorithm implementation process, wherein:
Fig. 3-the 1st, the user selectes client's window
Fig. 3-2 SIFT characteristics of image that algorithm extracts automatically in unrelated images window and photo 2
Fig. 3-3 algorithm finds the matching characteristic point in unrelated images window and the photo 2 automatically
Fig. 3-4 produces the replacement window by photo 2,3,4 and gives the user selection
Visitor on Fig. 3-5 step is removed from main photo
Remaining another one visitor in the selected main photo of Fig. 3-6 user
Fig. 3-7 produces the replacement window by photo 2,3,4 and gives the user selection
Fig. 4 is the end product that algorithm generates example image.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in detail, be to be noted that described embodiment only is intended to be convenient to the understanding of the present invention, and it is not played any qualification effect.
Shown in the process flow diagram among Fig. 1.At first take one group of photo by the user at floor, in this group photo, each part of background occurs once on a certain in the multiple photos of taking at least in the scene.In this group photo, by the user select one more clear and comprise irrelevant visitor or mobile object minimum as " main photo ".In main photo, specify rectangular window to comprise to want the irrelevant visitor who removes, i.e. " unrelated images window " by the user.After this, algorithm extracts topography's feature automatically in the photo of unrelated images window and other non-" main photo ", and carries out characteristic matching.Utilize the characteristics of image of coupling, algorithm computation goes out unrelated images window corresponding position in every non-main photo, and the image of these positions is generated a series of " replacement windows " through geometric transformation.For a unrelated images window, select an only replacement window by the user, algorithm replaces with the replacement window with the content of unrelated images window inside in the main photo in view of the above.According to this step, the user removes all the irrelevant visitors in the main photo successively one by one, just obtains an image that does not comprise any visitor at last.
Enforcement of the present invention is made up of two parts, i.e. and " photograph taking " and " photo processing ", as shown in Figure 1.Photograph taking uses digital camera to finish at floor by the user.These digital photographs import computing machine, move algorithm routine of the present invention on computers and carry out the photo processing.In the photo processing section, use a computer vision algorithm and a spot of user interactions of the present invention removed the visitor in the photo automatically.
In the photograph taking part, the user with identical angle, takes multiple photos according to certain time interval at same position for same scene.The time interval length of taking is held according to frequency and speed that irrelevant visitor in the scene moves by user oneself.The principle of taking is to make all scenery in the scene occur once on a certain in the multiple photos of taking at least.It may be noted that the present invention is also powerless to this so if certain part scenery all is in the state that is blocked by certain visitor (may not be same visitor) in all photos of taking, the user need be careful this situation and avoid as far as possible when taking.In shooting, the user can not use tripod, and hand held camera is taken and got final product.Though require the camera site of multiple image and angle to be consistent as far as possible, even there are deviation in the position and the angle of taking, the computer vision algorithms make that adopts among the present invention still can be revised it.Handle conveniently for follow-up photo, the suggestion user keeps the acquisition parameters of camera to fix (as aperture, focal length, shutter speed).
In the photo processing section, at first from this group multiple photos, select a more clear the most sparse photo of visitor that comprises simultaneously as " main photo " by the user.Then, the user specifies " unrelated images window " by mouse in main photo, and promptly one roughly comprises the rectangular window of wanting the irrelevant visitor that removes.The information of using for the image matching algorithm that guarantees to include in the unrelated images window abundant Gong back, actual unrelated images window is twice the expansion of the size of the rectangle of user's appointment, so just the picture material around the rectangle of user's appointment is also included as new unrelated images window.
The SIFT characteristics of image has partly been used in image characteristics extraction of the present invention, the SIFT feature is topography's feature description with very strong discriminating power, under conditions such as light intensity variation, image generation similarity transformation, still can find correct corresponding point reliably, with the comparison of other topography's descriptor in demonstrate outstanding performance.Use the SIFT feature can guarantee between two width of cloth images, to exist under the situation of camera site and angle difference, still can access correct coupling.In the characteristics of image compatible portion, algorithm uses arest neighbors and time neighbour's ratioing technigue to pick out reliable matching characteristic.When estimated image geometric transformation, thereby algorithm has used interative least square method to make algorithm robust more, and is insensitive for the match point of mistake.
In the feature extraction part, the present invention uses people's such as Lowe SIFT feature (Method andapparatus for identifying scale invariant features in an image and use of samefor locating an object in an image.David G.Lowe, US Patent 6,711,293 (March 23,2004) .Provisional application filed March 8,1999.Asignee:TheUniversity of British Columbia.).Algorithm extracts SIFT unique point and descriptor thereof from the unrelated images window, F is counted in its set
TWIf the photo of total m non-" main photo " is counted F from the SIFT characteristic set that wherein extracts
i, i ∈ 1,2 ..., m}.
At F
TWWith each F
iBetween seek the unique point set of coupling.For F
TWIn each SIFT feature P
j∈ F
TW, calculate P
jWith F
iIn the distance of each SIFT unique point, if the ratio of minimum distance and the second little distance is counted P so greater than threshold value 0.6
jBe match point, and write down it at F
iIn corresponding match point be Q
j, so just obtain F
TWAnd F
iBetween the set { P of matching characteristic point
j, Q
j}
(i)
Here we think that the geometric transformation between " unrelated images window " and the i photos concerns A
iSatisfy projective transformation, then an A
iIt is one 3 * 3 matrix (Ma Songde, Zhang Zhengyou, " computer vision ", Science Press, 1998).If F
TWIn the homogeneous coordinates of certain point be [u v1], F
iIn the homogeneous coordinates of certain point be [x y 1], then have: [u v 1]=[x y 1] * A
i
Utilize { P
j, Q
j}
(i), we use interative least square method can estimate A
iParameter (Li Qingyang etc., " numerical analysis ", publishing house of HUST, 1982).Utilize A
i, can calculate unrelated images window corresponding the window's position in the i photos, the picture material in this window is called in the i photos " the replacement window " to the unrelated images window.M width of cloth image will produce m and replace window.
If the background content of being blocked by the visitor in the unrelated images window occurs once on a certain in the multiple photos of taking at least, be the background that does not comprise any visitor with regard to having one content at least in this m replacement window so.Algorithm is replaced window with m and is shown to the user successively, selects a suitable replacement window by the user.
Algorithm uses the picture material in the selected replacement window to replace picture material in the unrelated images window, thereby this part the visitor in the main photo is removed.Equally, when the user continues to specify another unrelated images window, algorithm calculates once more according to above-mentioned process.The rest may be inferred till the user no longer continues to specify the unrelated images window.
Embodiment
Be one group of photo taking before church, Sofia, Harbin among Fig. 2, this is one of the significant sight spot in Harbin, and duration weekend, the visitor before the church is in an endless stream.The diverse location of visitor in picture all arranged in four photos among Fig. 2.In the present embodiment, the camera that we use is a Sony DSC T9 type digital camera, and shooting process is taken for hand-held.Note,, exist the different of position and angle between the multiple photos that handheld camera is taken owing to do not use tripod.The objective of the invention is to produce the photo that a width of cloth does not comprise any visitor by this four photos.
Four photos are directed in the computing machine and present to the user.At first pick out a main photo, the output photo that the algorithm of back will produce based on this by the user.Here, the user selects the photo 1 of visitor's minimum number as main photo.The user comprises the visitor on the step in the picture, shown in the interior rectangle among Fig. 3-1 by mouse selected rectangular window in main photo then.Outer rectangle among the figure is that algorithm is expanded this rectangle automatically with the background information around comprising.Picture material in the unrelated images window of outer rectangle representative is shown in Fig. 3-1.
Algorithm of the present invention extracts the SIFT feature automatically in unrelated images window and photo 2, photo 3, photo 4.With photo 2 is example, shown in Fig. 3-2.
With the SIFT feature in the unrelated images window respectively with photo 2, photo 3, photo 4 in the SIFT feature mate, find the corresponding relation between them, be example with photo 2, shown in Fig. 3-3, the same place among two width of cloth figure is marked with identical numeric character.
Use the unique point of coupling, algorithm estimates the position of unrelated images window in photo 2, photo 3, photo 4 by interative least square method, and the image of corresponding position process geometric transformation nothing to do with image window alignment generation replacement window, as shown in Figure 3-4, among the figure be the replacement window that produces in photo 2, photo 3, the photo 4 respectively.Can see at an easy rate, do not comprise any other visitor in the replacement window that has only photo 2 to produce, the user is selected with it, algorithm uses this to replace window and replaces the picture material in the current unrelated images window in the main photo, shown in Fig. 3-5, visitor in the unrelated images window is removed from picture, is come out by the intact recovery truly of background that he blocks simultaneously.
According to above-mentioned same process, the user continues the visitor on right side, gate in the selected picture, shown in Fig. 3-6.The replacement window that algorithm generates automatically is shown in Fig. 3-7.Result after the replacement as shown in Figure 4.At this moment in the photo without any the visitor, algorithm finishes.At last, we have obtained the photo that a width of cloth does not comprise any visitor, and church, Sofia in the photo background is real fully.
Describing above is to be used to realize embodiments of the invention, it should be appreciated by those skilled in the art, in any modification or partial replacement that does not depart from the scope of the present invention, all belongs to claim of the present invention and comes restricted portion.
Claims (4)
1. remove the method for unrelated images based on multiple photos for one kind, it is characterized in that step is as follows:
Step S1: the user takes multiple photos to a scene;
Step S2: choose and comprise the minimum photo of unrelated images in the multiple photos as main photo;
Step S3: judging has the unrelated images of denying to need to remove in the main photo, need to remove if having, and then execution in step 4; If do not need to remove, then execution in step 8;
Step S4: the selected unrelated images window that will remove unrelated images in main photo;
Step S5: utilize automated graphics registration technology matching image feature, calculate unrelated images window corresponding position in every non-main photo, and the image of these positions is generated a plurality of replacement windows through geometric transformation;
Step S6: in a plurality of replacement windows, select a replacement window that only comprises background;
Step S7: the picture material in the replacement window of selection is revised main photo through the picture material of replacing in the unrelated images window after the geometric transformation, and execution in step S3 revises main photo successively;
Step S8: successively all unrelated images in the main photo are removed one by one, obtained a background pictures that does not comprise unrelated images.
2. according to the method described in the claim 1, it is characterized in that:
Described multiple photos take be from a roughly the same background the position, according to the angle of roughly the same background, take at a certain time interval; In such multiple photos, occur once on the photo at least therein of each the part background in the scene.
3. according to the method described in the claim 1, it is characterized in that:
Described unrelated images window be shaped as rectangle, only need comprise unrelated images to small part, the size of this unrelated images window is enlarged to comprise the background information around the unrelated images automatically by algorithm.
4. according to the method described in the claim 1, it is characterized in that:
Described automated graphics registration uses the SIFT characteristics of image to carry out feature extraction and characteristic matching, and uses the geometric transformation parameter between interative least square method estimation unrelated images window and the replacement window.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100643896A CN101266685A (en) | 2007-03-14 | 2007-03-14 | A method for removing unrelated images based on multiple photos |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100643896A CN101266685A (en) | 2007-03-14 | 2007-03-14 | A method for removing unrelated images based on multiple photos |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101266685A true CN101266685A (en) | 2008-09-17 |
Family
ID=39989082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007100643896A Pending CN101266685A (en) | 2007-03-14 | 2007-03-14 | A method for removing unrelated images based on multiple photos |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101266685A (en) |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853402A (en) * | 2010-04-30 | 2010-10-06 | 北京航空航天大学 | Method for identifying barrier in perspective imaging process |
| CN102496147A (en) * | 2011-11-30 | 2012-06-13 | 宇龙计算机通信科技(深圳)有限公司 | Image processing device, image processing method and image processing system |
| WO2012075729A1 (en) * | 2010-12-08 | 2012-06-14 | 上海杰图软件技术有限公司 | Method and device for removing selected image in picture automatically |
| CN101482968B (en) * | 2008-01-07 | 2013-01-23 | 日电（中国）有限公司 | Image processing method and equipment |
| CN104079811A (en) * | 2014-07-24 | 2014-10-01 | 广东欧珀移动通信有限公司 | Method and device for filtering out obstacles during photographing |
| CN104092937A (en) * | 2014-06-16 | 2014-10-08 | 联想(北京)有限公司 | Image generation method and device |
| CN104104911A (en) * | 2014-07-04 | 2014-10-15 | 华中师范大学 | Timestamp eliminating and resetting method in panoramic image generation process and system thereof |
| CN104318251A (en) * | 2014-10-29 | 2015-01-28 | 广州视源电子科技股份有限公司 | Real-time processing method and device for photos |
| CN104349045A (en) * | 2013-08-09 | 2015-02-11 | 联想(北京)有限公司 | Image collecting method and electronic equipment |
| CN104539868A (en) * | 2014-11-24 | 2015-04-22 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN104580882A (en) * | 2014-11-03 | 2015-04-29 | 宇龙计算机通信科技(深圳)有限公司 | Photographing method and device |
| CN104735364A (en) * | 2013-12-19 | 2015-06-24 | 中兴通讯股份有限公司 | Photo shooting method and device |
| CN106210542A (en) * | 2016-08-16 | 2016-12-07 | 深圳市金立通信设备有限公司 | The method of a kind of photo synthesis and terminal |
| CN106331460A (en) * | 2015-06-19 | 2017-01-11 | 宇龙计算机通信科技(深圳)有限公司 | Image processing method and device, and terminal |
| CN107004296A (en) * | 2014-08-04 | 2017-08-01 | 脸谱公司 | For the method and system that face is reconstructed that blocks to reality environment |
| CN107395979A (en) * | 2017-08-14 | 2017-11-24 | 天津帕比特科技有限公司 | The image-pickup method and system of hollow out shelter are removed based on multi-angled shooting |
| CN107818538A (en) * | 2017-09-27 | 2018-03-20 | 百度在线网络技术（北京）有限公司 | Processing method, device, equipment and the computer-readable recording medium of watermarking images |
| CN107886471A (en) * | 2017-11-03 | 2018-04-06 | 北京航空航天大学 | A kind of unnecessary object minimizing technology of photo based on super-pixel Voting Model |
| CN108063884A (en) * | 2017-11-15 | 2018-05-22 | 维沃移动通信有限公司 | A kind of image processing method and mobile terminal |
| CN108259770A (en) * | 2018-03-30 | 2018-07-06 | 广东欧珀移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
| CN108447105A (en) * | 2018-02-02 | 2018-08-24 | 微幻科技(北京)有限公司 | A kind of processing method and processing device of panoramic picture |
| CN108494996A (en) * | 2018-05-14 | 2018-09-04 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and mobile terminal |
| CN108566516A (en) * | 2018-05-14 | 2018-09-21 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and mobile terminal |
| CN109167893A (en) * | 2018-10-23 | 2019-01-08 | Oppo广东移动通信有限公司 | Shoot processing method, device, storage medium and the mobile terminal of image |
| WO2019080061A1 (en) * | 2017-10-26 | 2019-05-02 | 深圳市柔宇科技有限公司 | Camera device-based occlusion detection and repair device, and occlusion detection and repair method therefor |
| CN109727266A (en) * | 2019-01-08 | 2019-05-07 | 青岛一舍科技有限公司 | A method of the target person photo based on the pure view background of video acquisition |
| CN110431835A (en) * | 2017-03-16 | 2019-11-08 | 富士胶片株式会社 | Image synthesizer, image composition method and program |
| CN110971825A (en) * | 2019-12-06 | 2020-04-07 | 深圳前海达闼云端智能科技有限公司 | Image correction method, electronic device and storage medium |
| CN112446820A (en) * | 2020-10-31 | 2021-03-05 | 浙江工业大学 | Method for removing irrelevant portrait of scenic spot photo |
| CN113014799A (en) * | 2021-01-28 | 2021-06-22 | 维沃移动通信有限公司 | Image display method and device and electronic equipment |
| CN114885086A (en) * | 2021-01-21 | 2022-08-09 | 华为技术有限公司 | Image processing method, head-mounted device and computer-readable storage medium |
-
2007
- 2007-03-14 CN CNA2007100643896A patent/CN101266685A/en active Pending
Cited By (45)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101482968B (en) * | 2008-01-07 | 2013-01-23 | 日电（中国）有限公司 | Image processing method and equipment |
| CN101853402B (en) * | 2010-04-30 | 2012-09-05 | 北京航空航天大学 | Method for identifying barrier in perspective imaging process |
| CN101853402A (en) * | 2010-04-30 | 2010-10-06 | 北京航空航天大学 | Method for identifying barrier in perspective imaging process |
| WO2012075729A1 (en) * | 2010-12-08 | 2012-06-14 | 上海杰图软件技术有限公司 | Method and device for removing selected image in picture automatically |
| CN102496147A (en) * | 2011-11-30 | 2012-06-13 | 宇龙计算机通信科技(深圳)有限公司 | Image processing device, image processing method and image processing system |
| CN104349045A (en) * | 2013-08-09 | 2015-02-11 | 联想(北京)有限公司 | Image collecting method and electronic equipment |
| CN104349045B (en) * | 2013-08-09 | 2019-01-15 | 联想(北京)有限公司 | A kind of image-pickup method and electronic equipment |
| CN104735364A (en) * | 2013-12-19 | 2015-06-24 | 中兴通讯股份有限公司 | Photo shooting method and device |
| CN104092937A (en) * | 2014-06-16 | 2014-10-08 | 联想(北京)有限公司 | Image generation method and device |
| CN104104911A (en) * | 2014-07-04 | 2014-10-15 | 华中师范大学 | Timestamp eliminating and resetting method in panoramic image generation process and system thereof |
| CN104104911B (en) * | 2014-07-04 | 2017-09-29 | 华中师范大学 | Timestamp in panoramic picture generating process is eliminated and remapping method and system |
| CN104079811A (en) * | 2014-07-24 | 2014-10-01 | 广东欧珀移动通信有限公司 | Method and device for filtering out obstacles during photographing |
| CN104079811B (en) * | 2014-07-24 | 2017-09-26 | 广东欧珀移动通信有限公司 | A kind of method and device that barrier is filtered out when taking pictures |
| CN107004296B (en) * | 2014-08-04 | 2020-12-29 | 脸谱科技有限责任公司 | Method and system for reconstructing an occlusion face of a virtual reality environment |
| CN107004296A (en) * | 2014-08-04 | 2017-08-01 | 脸谱公司 | For the method and system that face is reconstructed that blocks to reality environment |
| CN104318251A (en) * | 2014-10-29 | 2015-01-28 | 广州视源电子科技股份有限公司 | Real-time processing method and device for photos |
| CN104318251B (en) * | 2014-10-29 | 2018-05-01 | 广州视源电子科技股份有限公司 | Real-time processing method and device for photos |
| CN104580882A (en) * | 2014-11-03 | 2015-04-29 | 宇龙计算机通信科技(深圳)有限公司 | Photographing method and device |
| CN104580882B (en) * | 2014-11-03 | 2018-03-16 | 宇龙计算机通信科技(深圳)有限公司 | The method and its device taken pictures |
| CN104539868A (en) * | 2014-11-24 | 2015-04-22 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN104539868B (en) * | 2014-11-24 | 2018-06-01 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN106331460A (en) * | 2015-06-19 | 2017-01-11 | 宇龙计算机通信科技(深圳)有限公司 | Image processing method and device, and terminal |
| CN106210542A (en) * | 2016-08-16 | 2016-12-07 | 深圳市金立通信设备有限公司 | The method of a kind of photo synthesis and terminal |
| CN110431835B (en) * | 2017-03-16 | 2022-03-04 | 富士胶片株式会社 | Image synthesizing device, image synthesizing method, and recording medium |
| US11170477B2 (en) | 2017-03-16 | 2021-11-09 | Fujifilm Corporation | Image composition apparatus, image composition method, and program |
| CN110431835A (en) * | 2017-03-16 | 2019-11-08 | 富士胶片株式会社 | Image synthesizer, image composition method and program |
| CN107395979A (en) * | 2017-08-14 | 2017-11-24 | 天津帕比特科技有限公司 | The image-pickup method and system of hollow out shelter are removed based on multi-angled shooting |
| CN107818538A (en) * | 2017-09-27 | 2018-03-20 | 百度在线网络技术（北京）有限公司 | Processing method, device, equipment and the computer-readable recording medium of watermarking images |
| WO2019080061A1 (en) * | 2017-10-26 | 2019-05-02 | 深圳市柔宇科技有限公司 | Camera device-based occlusion detection and repair device, and occlusion detection and repair method therefor |
| CN107886471A (en) * | 2017-11-03 | 2018-04-06 | 北京航空航天大学 | A kind of unnecessary object minimizing technology of photo based on super-pixel Voting Model |
| CN108063884A (en) * | 2017-11-15 | 2018-05-22 | 维沃移动通信有限公司 | A kind of image processing method and mobile terminal |
| CN108447105A (en) * | 2018-02-02 | 2018-08-24 | 微幻科技(北京)有限公司 | A kind of processing method and processing device of panoramic picture |
| CN108259770B (en) * | 2018-03-30 | 2020-06-02 | Oppo广东移动通信有限公司 | Image processing method, image processing device, storage medium and electronic equipment |
| CN108259770A (en) * | 2018-03-30 | 2018-07-06 | 广东欧珀移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
| CN108566516A (en) * | 2018-05-14 | 2018-09-21 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and mobile terminal |
| CN108566516B (en) * | 2018-05-14 | 2020-07-31 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and mobile terminal |
| CN108494996B (en) * | 2018-05-14 | 2021-01-15 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and mobile terminal |
| CN108494996A (en) * | 2018-05-14 | 2018-09-04 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and mobile terminal |
| CN109167893B (en) * | 2018-10-23 | 2021-04-27 | Oppo广东移动通信有限公司 | Shot image processing method and device, storage medium and mobile terminal |
| CN109167893A (en) * | 2018-10-23 | 2019-01-08 | Oppo广东移动通信有限公司 | Shoot processing method, device, storage medium and the mobile terminal of image |
| CN109727266A (en) * | 2019-01-08 | 2019-05-07 | 青岛一舍科技有限公司 | A method of the target person photo based on the pure view background of video acquisition |
| CN110971825A (en) * | 2019-12-06 | 2020-04-07 | 深圳前海达闼云端智能科技有限公司 | Image correction method, electronic device and storage medium |
| CN112446820A (en) * | 2020-10-31 | 2021-03-05 | 浙江工业大学 | Method for removing irrelevant portrait of scenic spot photo |
| CN114885086A (en) * | 2021-01-21 | 2022-08-09 | 华为技术有限公司 | Image processing method, head-mounted device and computer-readable storage medium |
| CN113014799A (en) * | 2021-01-28 | 2021-06-22 | 维沃移动通信有限公司 | Image display method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101266685A (en) | A method for removing unrelated images based on multiple photos | |
| Jiang et al. | Cotr: Correspondence transformer for matching across images | |
| EP3182373B1 (en) | Improvements in determination of an ego-motion of a video apparatus in a slam type algorithm | |
| WO2017096949A1 (en) | Method, control device, and system for tracking and photographing target | |
| Kawewong et al. | Online and incremental appearance-based SLAM in highly dynamic environments | |
| Saxena et al. | Make3d: Learning 3d scene structure from a single still image | |
| Agarwal et al. | Building rome in a day | |
| KR102033262B1 (en) | Canonical reconstruction method, apparatus, terminal device and storage medium | |
| CN111382613B (en) | Image processing method, device, equipment and medium | |
| Sheng et al. | Unsupervised collaborative learning of keyframe detection and visual odometry towards monocular deep slam | |
| US9626585B2 (en) | Composition modeling for photo retrieval through geometric image segmentation | |
| CN101673346A (en) | Method, equipment and system for processing image | |
| CN108764269A (en) | A kind of cross datasets pedestrian recognition methods again based on space-time restriction incremental learning | |
| CN110599545B (en) | Feature-based dense map construction system | |
| EP2715613A2 (en) | Automatically optimizing capture of images of one or more subjects | |
| Krajník et al. | Image features and seasons revisited | |
| CN111323024A (en) | Positioning method and device, equipment and storage medium | |
| WO2019138983A1 (en) | Match determination device, match determination method, storage medium | |
| Zheng et al. | Sparse dynamic 3d reconstruction from unsynchronized videos | |
| JP2020161142A (en) | Mobile robot control method | |
| Mitra et al. | A large dataset for improving patch matching | |
| Zhang et al. | Cross-view image sequence geo-localization | |
| Morelli et al. | Photogrammetry now and then–from hand-crafted to deep-learning tie points– | |
| JP2017010549A (en) | Method and apparatus for data retrieval in lightfield database | |
| Ben-Artzi et al. | Event retrieval using motion barcodes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20080917 |