• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The present invention relates to a method for generating a cyclic video sequence (SC), characterized in that it comprises the implementation by data processing means (11) of a device (10) of steps of : (a) receiving from a video acquisition means (2) a video sequence (S); (b) Determining a first singular point (P1) of the video sequence (S) in a first interval (11) of said sequence (S) and a second singular point (P2) of the video sequence (S) in a second interval (12) of said sequence (S), said first and second singular points (P1, P2) having a maximum similarity according to a given similarity criterion. (c) Morphing generation of a coupling sequence (SR) between the image of the video sequence (S) corresponding to the second singular point (P1) and the image of the video sequence (S) corresponding to the first singular point (P2); (d) constructing said cyclic video sequence (SC) by concatenating a fragment (S ') of said video sequence (S) extending from the first singular point (P1) to the second singular point (P2), and said connecting sequence (SR).
    公开号:FR3033919A1
    申请号:FR1552350
    申请日:2015-03-20
    公开日:2016-09-23
    发明作者:Thomas Niel Vassort
    申请人:Thomas Niel Vassort；
    IPC主号:
    专利说明:

    [0001] TECHNICAL FIELD The present invention relates to the field of video animation. More specifically, it relates to a method for generating a cyclic video sequence. STATE OF THE ART It is now common to hear "cyclic" sound sequences, that is to say, played in a loop: the sequence is designed to present an identical beginning and end, so that it It is possible to repeat it over and over a large number of times, with a listener having the impression that it is a continuous sequence. The realization of a "cyclic" video sequence is much more complex, insofar as a set of actors is controlled much less well than a sound, and it is impossible to achieve an exactly identical beginning and end. Such a sequence played in a loop can not deceive the human eye, which instantly sees the "connection", that is to say the moment when we restart the reading.
    [0002] Thus, the sequences played for example on wall advertising media are generally clips separated by black fades. If we still want to give the illusion of a loop, we know the animated GIFs, which consist of a succession of images, but not a real video sequence. Since the sequence is much less than 24 frames per second, there is often a significant gap between two successive images and the human eye is much more "tolerant" to the connection. The animated GIF is often used for what are called "cinemagraphs", that is to say animated photographs of a slight repetitive movement (usually on a small portion in the foreground of the image, the background remaining fixed ). The possibilities offered by the animated GIF 3033919 2 however remain very limited, and in all cases only allow a mediocre quality far from "film" quality video sequences. It would therefore be desirable to have a solution allowing a user to make movie-quality looped video sequences, with a rendering such that the human eye feels like a continuous sequence. PRESENTATION OF THE INVENTION The present invention thus relates in a first aspect to a method for generating a cyclic video sequence, characterized in that it comprises the implementation by data processing means of a device steps of: (a) receiving from a video acquisition means a video sequence; (b) determining a first singular point of the video sequence in a first interval of said sequence and a second singular point of the video sequence in a second interval of said sequence, said first and second singular points having a similarity maximum according to a given similarity criterion; (c) Morphing generation of a coupling sequence between the image of the video sequence corresponding to the second singular point and the image of the video sequence corresponding to the first singular point; (D) constructing said cyclic video sequence by concatenating a fragment of said video sequence extending from the first singular point to the second singular point, and said coupling sequence.
    [0003] From a standard filmed video sequence, the present invention combines the determination of a pair of optimal singular points to define a quasi-cyclic sequence body extending from the first to the second singular point, and the generation by morphing a coupling sequence to return to the first singular point from the second singular point. Thus the final sequence obtained is completely cyclic and can be looped with a perfect rendering since its beginning and end coincide exactly. According to other advantageous and non-limiting characteristics: said video sequence comprises a substantially periodic pattern of a given duration less than the duration of the video sequence, the second interval corresponding to the first interval offset by a duration approximately equal to one multiple of said pattern duration; said video sequence comprises several substantially periodic patterns, the method comprising performing steps (b) to (d) for each of the patterns so as to generate a plurality of cyclic sequences; the method comprises a step (e) of constructing at least one transition sequence between two of the generated cyclic sequences, said transition sequence being a fragment of said video sequence extending from the first singular point of one of said two cyclic sequences at the first singular point of the other of said two cyclic sequences; said first interval defines a subsequence of said video sequence having a more dynamic character than the rest of the video sequence; the evaluation of the dynamic nature of a point of the sequence also comprises the comparison of the image of the sequence corresponding to the point with the preceding image and / or the next image of the sequence; Said dynamic character relates to at least one area of attention of the image; Step b) comprises for each pair of a point of the first interval and a point of the second interval the evaluation of the similarity between the two points, the first and second singular points being chosen as the pair points with the highest similarity; Said criterion of similarity between two points of the video sequence comprises a static component and a dynamic component; the evaluation of the static component of the similarity criterion between two points of the sequence comprises the comparison of the two images of the sequence corresponding to each of the two points; The evaluation of the dynamic component of the criterion of similarity between two points of the sequence comprises comparing the two preceding images and / or the two images according to the two images of the sequence corresponding to each of the two points; the evaluation of the dynamic component of the criterion of similarity between two points of the sequence also comprises the respective comparison of each of the two images of the sequence corresponding to the two points with the preceding image and / or the following image of the sequence. According to a second aspect, the invention relates to a system for generating a cyclic video sequence, comprising means for video acquisition of a video sequence and equipment comprising data processing means configured to implement: A module for receiving said video sequence; A module for determining a first singular point of the video sequence in a first interval of said sequence and a second singular point of the video sequence in a second interval of said sequence, said first and second singular points exhibiting a maximum similarity according to a given similarity criterion; A module for generating by morphing a coupling sequence between the image of the video sequence corresponding to the second singular point and the image of the video sequence corresponding to the first singular point; A module for constructing said cyclic video sequence by concatenating a fragment of said video sequence extending from the first singular point to the second singular point, and said connecting sequence. According to a third and fourth aspect, the invention relates respectively to a computer program product comprising code instructions for executing, when said program product is executed by data processing means, a method according to the first aspect of the invention generating a cyclic video sequence; and computer readable storage means on which a computer program product comprises code instructions for executing a method according to the first aspect of the invention for generating a cyclic video sequence. PRESENTATION OF THE FIGURES Other characteristics and advantages of the present invention will appear on reading the following description of a preferred embodiment. This description will be given with reference to the accompanying drawings, in which: FIG. 1 represents an embodiment of a system for implementing a method according to the invention; FIG. 2 illustrates the generation of a cyclic sequence thanks to an embodiment of the method according to the invention. DETAILED DESCRIPTION Architecture 3033 919 6 Referring to FIG. 1, the present method of generating a cyclic video sequence is implemented at a device 10 comprising data processing means 11 (in particular a processor ) and data storage means 12 (a memory, for example a hard disk). The equipment 10 is typically a workstation, equipped with a user interface 13 (a keyboard, a mouse, a screen, etc.). The equipment 10 is connected directly or indirectly to video acquisition means 2. These latter typically consist of one or more conventional cameras capable of acquiring a video sequence S representing a scene 1. By "directly or indirectly", we mean: in the first case that the means 2 and the equipment 10 are connected during data acquisition via a cable or a network, the video file produced during the acquisition can be loaded in real time on equipment 10; in the second case, the means 2 comprise their own memory, which is connected in a second step to the equipment 10 for loading the video sequence on the data storage means 12 for processing. In any case, it will be understood that the present method is not limited to any mode of operation of the acquisition means 2, it is only necessary that they be capable of generating a digital file representative of a video sequence S, so that the data processing means 11 of the equipment 10 can receive them in a step (a). In most cases, the video sequence S will be stored on the storage means 12 for processing. Note that in most cases the acquisition means 2 will be fixed in front of a scene 1 having some moving elements, which will facilitate the implementation of the method and improve its rendering.
    [0004] The present method is directed to generating a cyclic sequence SC from an initial video sequence S.
    [0005] We note that here we are in the case of a video sequence S "real", that is to say a scene 1 of reality that can be filmed by a video camera, and not images of synthesis for example. The cyclic sequence SC is a sequence capable of being played in a loop (unlike the sequence S), that is to say that it has a beginning and an end sufficiently close to make it possible to repeat it without that the human eye is not able (or very difficult) to detect the restart. The initial sequence S can be any sequence, although it is naturally preferred (for a question of quality of the final rendering) that it contains a substantially periodic pattern, and that it is generally fixed, for example filmed. on a green background, to improve the "homogeneity" of the sequence. To return to the periodic aspect, it is desirable, for example, for this sequence S to represent an actor continuously moving the same movement several times. In other words, the actor returns one or more times substantially to its starting point during movement. At this level, it is not necessary (and, moreover, impossible) for the actor to make the same movement several times, it is sufficient for him to pass through intermediate positions sufficiently close so that "singular points" can be identified. Thus, by pattern, it is meant that the sequence S attempts to simulate a loop by including an element approximately reproduced one or more times. As an alternative to the presence of patterns, it will be noted that the sequence may be of the "animated image" type, that is to say take the form of a generally fixed scene that takes a large number of variations. For example, you can film a person's face by asking them to take a large number of expressions in a row. The case of the animated image 03 03 3 9 1 9 8 will be treated more particularly later, we will for the moment use the example of the periodic pattern. Singular points With reference to FIG. 2, the method comprises in a step (b) the determination of a first singular point P1 of the video sequence S in a first interval 11 of said sequence S and a second singular point P2 of the video sequence S in a second interval 12 of said sequence S, said first and second singular points Pi, P2 having a maximum similarity according to a given similarity criterion. The idea is to identify the two points of the sequence S which are the most "similar", that is to say the most suitable to serve as a point of closure (that is to say of beginning / end of the loop).
    [0006] These singular points are each searched for over a probable interval, i.e. the first and second intervals 11, 12 correspond to fragments of the S sequence more likely than others to comprise singular points. It is noted that one and / or the other of the first and second intervals 11, 12 may be the entire S sequence, but this greatly slows down the implementation of the method. Preferably, the two intervals 11, 12 are disjoint, and more particularly the second interval 12 is defined as a function of the first interval 11. In fact, logically the two most similar points of any video sequence are two immediately adjacent points (separated from an image), which is not of interest for generating the cyclic sequence SC. It is desirable that it has a minimum length, hence the interest of defining intervals. Preferably, if there is an estimate of the duration of a pattern, the second interval corresponds to the first offset interval of the same duration (or a multiple of the same duration): the two intervals 11 , 12 thus correspond to very similar fragments of the sequence S, which are likely to contain similar points. It should be noted that the intervals are not necessarily 3033919 9 continuous. In particular, if a pattern is repeated more than once, the second interval 12 may be defined as the union of the offsets of the first interval 11 by at least once the duration of the pattern. The first interval 11 is not necessarily chosen at the beginning of a pattern: it may be in the middle, the second interval 12 then being at the same level in the next iteration of the pattern, or in an even later iteration (ie offset by a multiple of the approximate duration of the reason, "multiple" including once). Preferably, the first interval 11 is chosen to correspond to a moment of the pattern having a "dynamic" character, that is to say present locally a strong movement, in contrast with a rather static moment. In a preferred manner, the dynamism concerns so-called "attention" zones of the image. These are points traditionally attracted to the human eye such as a face, a hand, etc. These so-called attention zones will be chosen preferentially as parts of animated beings (humans, animals) visible in the sequence S. These areas of attention are indeed those at which: - on the one hand deviations are the more likely to exist, since by definition it is the "living" movements that can not be reproduced exactly identically, unlike those of inanimate objects; and on the other hand a human will be able to easily notice a bad connection. The more they move, the more the loop will go unnoticed.
    [0007] Areas of attention may be identified by image recognition or designated by a user. Examples of ways of quantifying the dynamic nature of a point in the sequence will be defined in the remainder of this description. The length of the gaps 11, 12 is chosen according to the length of the sequence S, its dynamism, the power of the data processing means 11, etc. A few seconds 3033919 10 appear sufficient. It should be noted that the intervals can be defined by an operator on the interface 13. The example of FIG. 2 represents an example of a sequence S comprising three patterns. In other words, the filmed object passes four times (approximately each time) by its starting point. The moment of strongest dynamism is about one third of motive, we define the first interval at this level. The second interval 12 is defined as the translation of the first interval 11 of one and / or two patterns. In the example shown, we only take the one shifted once.
    [0008] Similarity By "similarity" is meant the level of similarity for the human eye. The greater the similarity between two images, the more a human 15 can believe that it is the same image, and not see any difference between these images. As explained, the two points having the maximum similarity are identified in each interval 11, 12. It may be a raw iterative test comprising for each pair a point of the first interval 11 and a point of the second interval 12, the evaluation of a similarity between the two points, the first and second singular points being chosen as the pair of points with the highest similarity. Alternatively, it is sufficient to find two points with a similarity beyond an acceptable threshold. Note that this step can be performed purely by the data processing means 11, or include the verification of a user who will compare pairs of candidate points and designate the best in the sense of the human eye. FIG. 1 represents for example on the interface means 13 selected images of each of the intervals 11, 12, the operator making the final choice.
    [0009] The similarity evaluation between two points of the sequence S preferably comprises the calculation of two components: a "static" component, i.e. a comparison, in particular pixel by pixel, of the two target images. For example, a score can represent the number of different pixels, or the average difference in RGB pixels two by two, etc. It should be noted that the importance of the so-called attention points mentioned above can be taken into account by assigning them a higher weighting coefficient; a "dynamic" component. Indeed, it is not impossible that a sequence includes two identical images, but not connectable. For example, if we film a pendulum swinging from left to right, on a pattern it will pass twice exactly in the center (once from the left and once from the right). These two images may be perfectly identical, they are not similar because the movement is not the same (once the pendulum comes from the left and the other time from the right), and a human eye would see everything from a problem if we presented them chained. For this reason, the dynamic component is evaluated for example by looking at the images before and / or after the target image. It is desirable: That the images immediately before (respectively immediately after) each of the target images are the most identical to each other "statically" (it will be possible to use the criteria defined above), which makes it possible to show that the motion Is similar. Indeed, if the two target images are identical but not the two images just before, they are each part of a different movement and they are not acceptable for a connection; That the image immediately before (respectively after) each of the target images and the target image itself are the most different, i.e., the least identical (the comparison mechanisms mentioned just before) can be used. Indeed, the more the two successive images are different, the more the dynamic nature of the moment is pronounced, and more a connection at this time will be imperceptible. The comparison of two successive images is thus once to evaluate the dynamic character of a point of the sequence (see above). It is also possible to apply higher weighting coefficients to the areas of attention. Thus the similarity can be calculated for example as a linear combination of its static and dynamic scores. In general, it will be understood that the singular points P1 and P2 most often correspond to close target images preceded or followed by images close to each other, but quite different from the target images themselves. It should be understood that the present invention is not limited to any particular way of evaluating the similarity of points in the S-sequence, and those skilled in the art will be able to transpose a large number of known image analysis techniques. Morphing At the end of step (b), there are first and second singular points Pi, P2 of the sequence, which are two of the most similar points of the sequence. The fragment S 'of said video sequence S extending from the first singular point P1 to the second singular point P2 (see FIG. 2) could almost be looped. Nevertheless, these points Pi, P2 are never similar to 100 ° A, and a loop between P2 and P1 would be visible to the human eye that would see the transition. The present method thus comprises, in a step (c), the generation of a connection sequence SR between the image of the video sequence S corresponding to the second singular point P1 and the image of the video sequence (S) corresponding to the first singular point P2. This SR coupling sequence is performed by morphing (or "morphose"), a well-known technique. Since the points P2 and P1 are very similar, the morphing is slight and goes unnoticed by the spectators. The duration of the coupling sequence is advantageously very short, typically less than one second. It should be noted that the length of the connection sequence SR can be a function of the dynamic nature of the sequence S at the points P1 and P2: the more dynamic the movement, the shorter the connection sequence SR must be in order not to break the movement. The skilled person can use any existing algorithm for the implementation of this morphing step.
    [0010] Construction of the cyclic sequence In a last step (d), the data processing means 11 builds the cyclic video sequence SC by concatenating the fragment S 'of said video sequence S extending from the first singular point P1 to the second point singular P2, and said connection sequence SR. Thus, during the sequence S ', one goes from P1 to P2, and during the connection sequence, one goes from P2 to P1, which closes the loop.
    [0011] The beginning and the end of the cyclic sequence SC obtained are thus exactly identical, and the sequence SC can be broadcast in a perfectly fluid loop and chained on display means (such as a wall-mounted advertising medium), and the observer will not see the transition.
    [0012] Case of the animated image As explained, the animated image corresponds to a globally fixed scene 1, animated by small, almost random movements (movements of vegetation in the wind, expression of the face of an actor, 30 etc.). Such a video sequence S does not comprise patterns, but has a large number of potential connection points (as the image 3033919 14 is globally fixed, with no overall motion). We note that the dynamic character is globally constant. The first gap 11 may be arbitrarily defined, and the second gap 12 may be an offset of the first by a length corresponding to the minimum length of the desired loop. By prolonging the acquisition of the sequence S long enough, the chances of finding a near-perfect loop point increase sharply, step (b) can then be implemented by simple iteration, until two points are found. and P2 having a similarity beyond a predefined threshold. Multiple or dynamic loops According to a particular embodiment (and if the sequence S allows it), it is possible to generate multiple SC cyclic sequences. The idea is to identify several patterns within the initial S sequence and define for each a loop. For example, the actor can, continuously, make a movement A three times (first pattern) then twice a movement B (second pattern). Steps (b) to (d) can then be implemented for each of the patterns so as to generate several cyclic sequences SC which can each be broadcast in a loop, while allowing transitions from one cyclic sequence to another cyclic sequence. . At least one of these transitions can be constructed in a step (e) by extracting the S fragment binding a singular point of a cyclic sequence SC at a singular point of another. For example, if a first cyclic sequence SCa is defined by a fragment [P1-P2] of S (between the singular points P1 and P2) and a second cyclic sequence SCb is defined by a fragment [P3-P4] of S between singular points P3 and P4, with the points P1, P3, P2 and P4 being arranged in this order within the sequence S, the fragment [Pl-P3] defines a transition from SCa to SCb, and the fragment [P3- P2] + SRa (the coupling sequence associated with the first cyclic sequence SCa, corresponding to the morphing of P2 to P1) defines a transition from SCb to SCa. Provided that a very long S sequence is available and comprises numerous patterns (or even several S sequences in which the actor 5 performs the same movements), it is possible to generate a family of cyclic sequences SC, the user being able for example to trigger transitions from one to another depending on various interactions when it is broadcast (for example by touching an area of the display, triggering a motion detector past the display, etc.). According to another particular embodiment, the loop can be dynamic, and change from one iteration to another (or after a certain number of iterations), either by processing the cyclic sequence SC so as to obtain modified cyclic sequences SC ', SC ", etc., using the multiple loops technique hereinbefore, for example, a loop whose duration follows a dynamic element, for example the seasons. the loop will change from day to day with the length of the day, and will change completely each season (a winter buckle, an autumn buckle, a summer buckle, and a spring buckle.) System & Computer Program Product According to a second aspect The invention relates to a system for generating a cyclic video sequence SC for implementing the method according to the first aspect, the system comprising, as explained, video acquisition means 2 of a video sequence S and a equipment 10 comprising data processing means 11 (and typically data storage means 12 and an interface 13). The data processing means 11 are configured to implement: a module for receiving said video sequence S; A module for determining a first singular point P1 of the video sequence S in a first interval 11 of said sequence S and a second singular point P2 of the video sequence S in a second interval 12 of said sequence S, said first and second singular points Pi, P2 having a maximum similarity according to a given similarity criterion; A morphing generation module of a connection sequence SR between the image of the video sequence S corresponding to the second singular point P1 and the image of the video sequence S corresponding to the first singular point P2; A module for constructing said cyclic video sequence SC by concatenating a fragment S 'of said video sequence S extending from the first singular point P1 to the second singular point P2, and said connection sequence SR. According to a third and fourth aspect, the invention relates to a computer program product comprising code instructions for execution, when this program product is executed by data processing means 11 (typically those of the equipment). 10), a method according to the first aspect of the invention for generating a cyclic video sequence, as well as storage means readable by a computer equipment (typically the data storage means 12 of the equipment 10) on which a computer program product on which these code instructions are found.
    权利要求:
    Claims (15)
    [0001]
    REVENDICATIONS1. A method of generating a cyclic video sequence (SC), characterized in that it comprises the implementation by data processing means (11) of a device (10) of steps of: (a) Reception from video acquisition means (2) of a video sequence (S); (b) Determining a first singular point (P1) of the video sequence (S) in a first interval (11) of said sequence (S) and a second singular point (P2) of the video sequence (S) in a second interval (12) of said sequence (S), said first and second singular points (P1, P2) having a maximum similarity according to a given similarity criterion; (c) Morphing generation of a coupling sequence (SR) between the image of the video sequence (S) corresponding to the second singular point (P1) and the image of the video sequence (S) corresponding to the first singular point (P2); (d) constructing said cyclic video sequence (SC) by concatenating a fragment (S ') of said video sequence (S) extending from the first singular point (P1) to the second singular point (P2), and said connecting sequence (SR).
    [0002]
    The method of claim 1, wherein said video sequence (S) comprises a substantially periodic pattern of a given duration less than the duration of the video sequence (S), the second interval (12) corresponding to the first interval (11). ) offset by a duration approximately equal to a multiple of the said duration of the pattern.
    [0003]
    The method of claim 2, wherein said video sequence (S) comprises a plurality of substantially periodic patterns, the method comprising performing steps (b) through (d) for each of the patterns so as to generate a plurality cyclic sequences (SC).
    [0004]
    4. The method according to claim 3, comprising a step (e) of constructing at least one transition sequence between two of the generated cyclic sequences (SC), said transition sequence being a fragment of said video sequence (S). extending from the first singular point (P1) of one of said two cyclic sequences (SC) to the first singular point (P1) of the other of said two cyclic sequences (SC). 10
    [0005]
    5. Method according to one of claims 1 to 4, wherein said first interval (11) defines a subsequence of said video sequence (S) having a more dynamic character than the rest of the video sequence (S). 15
    [0006]
    The method according to claim 5, wherein the evaluation of the dynamic character of a point of the sequence (S) also comprises comparing the image of the sequence (S) corresponding to the point with the preceding image and / or the next image of the sequence (S).
    [0007]
    7. Method according to one of claims 5 and 6, wherein said dynamic character relates to at least one area of attention of the image. 25
    [0008]
    The method according to one of claims 1 to 7, wherein step (b) comprises for each pair of a point of the first interval (11) and a point of the second interval (12) the evaluation of the similarity between the two points, the first and second singular points being chosen as the pair of points having the highest similarity. 3033919 19
    [0009]
    9. Method according to one of claims 1 to 8, wherein said criterion of similarity between two points of the video sequence comprises a static component and a dynamic component. 5
    [0010]
    10. The method of claim 9, wherein the evaluation of the static component of the similarity criterion between two points of the sequence (S) comprises comparing the two images of the sequence (S) corresponding to each of the two points. 10
    [0011]
    11. Method according to one of claims 9 and 10, wherein the evaluation of the dynamic component of the criterion of similarity between two points of the sequence (S) comprises the comparison of the two preceding images and / or two images according to the two images of the sequence (S) corresponding to each of the two points. 15
    [0012]
    The method according to claim 11, wherein the evaluation of the dynamic component of the similarity criterion between two points of the sequence (S) also comprises the respective comparison of each of the two images of the sequence (S) corresponding to both of them. points with the previous image and / or the next image of the sequence (S).
    [0013]
    13. System for generating a cyclic video sequence (SC), comprising means for video acquisition (2) of a video sequence (S) and equipment (10) comprising data processing means (11) configured to implement: a module for receiving said video sequence (S); A module for determining a first singular point (P1) of the video sequence (S) in a first interval (11) of said sequence (S) and a second singular point (P2) of the video sequence ( S) in a second interval (12) of said sequence (S), said first and second singular points (P1, P2) having a maximum similarity according to a given similarity criterion; A morphing generation module of a connection sequence (SR) between the image of the video sequence (S) corresponding to the second singular point (P1) and the image of the video sequence (S) corresponding to the first singular point (P2); A module for constructing said cyclic video sequence (SC) by concatenating a fragment (S ') of said video sequence (S) extending from the first singular point (P1) to the second singular point (P2), and of said connecting sequence (SR). 10
    [0014]
    14. A computer program product comprising code instructions for executing, when this program product is executed by data processing means (11), a method according to one of claims 1 to 12 of the present invention. a cyclic video sequence (SC). 15
    [0015]
    15. Storage medium readable by a computer equipment on which a computer program product comprises code instructions for the execution of a method according to one of claims 1 to 12 for generating a cyclic video sequence (SC ).
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3070678B1|2020-10-07|Method for generating a cyclic video sequence
    EP2393022A2|2011-12-07|Method for creating a media sequence using coherent groups of media files
    EP2491532A1|2012-08-29|Method, computer program, and device for hybrid tracking of real-time representations of objects in image sequence
    WO2016071577A1|2016-05-12|Method for producing animated images
    WO2013088076A1|2013-06-20|System for filming a video movie
    WO2017001768A1|2017-01-05|Device for processing data for additive manufacturing
    FR2996034A1|2014-03-28|Method for generating high dynamic range image representing scene in e.g. digital still camera, involves generating composite images by superposition of obtained images, and generating high dynamic range image using composite images
    EP3210378B1|2020-11-04|Method for collecting image data for producing immersive video and method for viewing a space on the basis of the image data
    EP3918576A1|2021-12-08|Dynamic three-dimensional imaging method
    US11182618B2|2021-11-23|Method and system for dynamically analyzing, modifying, and distributing digital images and video
    FR2834854A1|2003-07-18|METHOD AND SYSTEM FOR TATTOO DECODING
    FR3038253A1|2017-01-06|DATA PROCESSING DEVICE
    EP2626837A1|2013-08-14|System for creating three-dimensional representations from real models having similar and pre-determined characteristics
    FR3047825B1|2019-10-11|METHOD FOR IDENTIFYING A SHOW IN A VIDEO FILMED BY A CAMERA OF A SPECTATOR
    FR3105687A1|2021-06-25|PROCESS FOR ENCODING A DIGITAL IMAGE FOR COMPRESSION
    Zhang2021|Bridging Machine Learning and Computational Photography to Bring Professional Quality into Casual Photos and Videos
    CA2978822A1|2016-09-15|Image processing device
    FR3043233A1|2017-05-05|METHOD OF AUTOMATICALLY SELECTING IMAGES FROM A MOBILE DEVICE
    Goyal et al.2010|Image Inpainting
    FR2817625A1|2002-06-07|Method for detecting objects in an illuminated area, such as a person under water, uses digitized video images
    WO2012004537A1|2012-01-12|Image synthesis device
    WO2007028892A1|2007-03-15|Method for monitoring at least one deformable object in a sequence of images, corresponding system and computer programme
    FR2823630A1|2002-10-18|MPEG watermarking system has segment data in signal avoids separate file
    同族专利:
    公开号 | 公开日
    EP3070678B1|2020-10-07|
    FR3033919B1|2018-09-07|
    US20160275990A1|2016-09-22|
    US9852767B2|2017-12-26|
    EP3070678A1|2016-09-21|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP2557773A1|2011-03-18|2013-02-13|Sony Corporation|Image processing device and method; and program|
    US6504990B1|1998-11-12|2003-01-07|Max Abecassis|Randomly and continuously playing fragments of a video segment|
    US6408128B1|1998-11-12|2002-06-18|Max Abecassis|Replaying with supplementary information a segment of a video|
    US6611268B1|2000-05-30|2003-08-26|Microsoft Corporation|System and process for generating 3D video textures using video-based rendering techniques|
    US6636220B1|2000-01-05|2003-10-21|Microsoft Corporation|Video-based rendering|
    US20060129933A1|2000-12-19|2006-06-15|Sparkpoint Software, Inc.|System and method for multimedia authoring and playback|
    US20050022252A1|2002-06-04|2005-01-27|Tong Shen|System for multimedia recognition, analysis, and indexing, using text, audio, and digital video|
    US20120276509A1|2010-10-29|2012-11-01|The Cleveland Clinic Foundation|System of preoperative planning and provision of patient-specific surgical aids|
    US8872850B2|2012-03-05|2014-10-28|Microsoft Corporation|Juxtaposing still and dynamic imagery for cliplet creation|
    US9141281B2|2012-09-28|2015-09-22|Fabrizio Ferri Photography|System and method for controlling the progression of multmedia assets on a computing device|
    US20160104307A1|2014-10-14|2016-04-14|Microsoft Technology Licensing, Llc.|Data visualization extensibility architecture|US10845955B2|2017-05-15|2020-11-24|Apple Inc.|Displaying a scrollable list of affordances associated with physical activities|
    DK201870599A1|2018-03-12|2019-10-16|Apple Inc.|User interfaces for health monitoring|
    DK201870380A1|2018-05-07|2020-01-29|Apple Inc.|Displaying user interfaces associated with physical activities|
    US11228835B2|2019-06-01|2022-01-18|Apple Inc.|User interfaces for managing audio exposure|
    US11209957B2|2019-06-01|2021-12-28|Apple Inc.|User interfaces for cycle tracking|
    US11234077B2|2019-06-01|2022-01-25|Apple Inc.|User interfaces for managing audio exposure|
    DK201970534A1|2019-06-01|2021-02-16|Apple Inc|User interfaces for monitoring noise exposure levels|
    US11152100B2|2019-06-01|2021-10-19|Apple Inc.|Health application user interfaces|
    WO2021051121A1|2019-09-09|2021-03-18|Apple Inc.|Research study user interfaces|
    US20210373748A1|2020-06-02|2021-12-02|Apple Inc.|User interfaces for health applications|
    法律状态:
    2016-03-24| PLFP| Fee payment|Year of fee payment: 2 |
    2016-09-23| PLSC| Publication of the preliminary search report|Effective date: 20160923 |
    2017-03-09| PLFP| Fee payment|Year of fee payment: 3 |
    2018-03-27| PLFP| Fee payment|Year of fee payment: 4 |
    2020-03-31| PLFP| Fee payment|Year of fee payment: 6 |
    2021-04-29| PLFP| Fee payment|Year of fee payment: 7 |
    2022-02-11| PLFP| Fee payment|Year of fee payment: 8 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1552350A|FR3033919B1|2015-03-20|2015-03-20|METHOD FOR GENERATING A CYCLIC VIDEO SEQUENCE|
    FR1552350|2015-03-20|FR1552350A| FR3033919B1|2015-03-20|2015-03-20|METHOD FOR GENERATING A CYCLIC VIDEO SEQUENCE|
    EP16161278.3A| EP3070678B1|2015-03-20|2016-03-18|Method for generating a cyclic video sequence|
    US15/074,645| US9852767B2|2015-03-20|2016-03-18|Method for generating a cyclic video sequence|
    [返回顶部]