WO2022215823A1 - 영상 생성 방법 및 장치 - Google Patents
영상 생성 방법 및 장치 Download PDFInfo
- Publication number
- WO2022215823A1 WO2022215823A1 PCT/KR2021/017653 KR2021017653W WO2022215823A1 WO 2022215823 A1 WO2022215823 A1 WO 2022215823A1 KR 2021017653 W KR2021017653 W KR 2021017653W WO 2022215823 A1 WO2022215823 A1 WO 2022215823A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- character
- shot
- determining
- image
- movement
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 230000001815 facial effect Effects 0.000 claims description 76
- 230000008859 change Effects 0.000 claims description 30
- 230000008921 facial expression Effects 0.000 claims description 30
- 230000003993 interaction Effects 0.000 claims description 28
- 230000014509 gene expression Effects 0.000 claims description 16
- 210000004709 eyebrow Anatomy 0.000 claims description 12
- 210000001747 pupil Anatomy 0.000 claims description 11
- 230000007704 transition Effects 0.000 claims description 11
- 210000003128 head Anatomy 0.000 claims description 8
- 210000000744 eyelid Anatomy 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000010304 firing Methods 0.000 claims description 4
- 230000004044 response Effects 0.000 description 22
- 238000004891 communication Methods 0.000 description 21
- 230000036544 posture Effects 0.000 description 14
- 238000012545 processing Methods 0.000 description 11
- 230000015654 memory Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 230000008451 emotion Effects 0.000 description 4
- 230000005291 magnetic effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 230000006855 networking Effects 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241001300198 Caperonia palustris Species 0.000 description 1
- 241000556720 Manga Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 235000000384 Veronica chamaedrys Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 210000000617 arm Anatomy 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 210000003127 knee Anatomy 0.000 description 1
- 210000002414 leg Anatomy 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000214 mouth Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004091 panning Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B27/00—Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
- G11B27/02—Editing, e.g. varying the order of information signals recorded on, or reproduced from, record carriers
- G11B27/031—Electronic editing of digitised analogue information signals, e.g. audio or video signals
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B27/00—Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
- G11B27/02—Editing, e.g. varying the order of information signals recorded on, or reproduced from, record carriers
- G11B27/031—Electronic editing of digitised analogue information signals, e.g. audio or video signals
- G11B27/036—Insert-editing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T13/00—Animation
- G06T13/20—3D [Three Dimensional] animation
- G06T13/40—3D [Three Dimensional] animation of characters, e.g. humans, animals or virtual beings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/04—Segmentation; Word boundary detection
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/183—Speech classification or search using natural language modelling using context dependencies, e.g. language models
- G10L15/187—Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
- G10L21/10—Transforming into visible information
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/57—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for processing of video signals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Definitions
- the following embodiments relate to a method and apparatus for generating an image, and more particularly, to a method and apparatus for generating an image including a character.
- SNS social networking service
- SNS is an online platform service for interacting with other users through a network.
- SNS provides social networking functions through the exchange of information in various forms such as texts, photos, videos, and voices.
- technologies for various types of social networking services that can freely express their individuality and thoughts and communicate with other users conveniently.
- the following embodiments may provide a technique for generating a moving character according to a user's input and generating an image obtained by photographing the moving character using various shooting techniques.
- the following embodiments may provide an eternal life generation technology that automatically generates elements necessary for image generation and provides an editing function of the automatically generated elements.
- An image generating method includes: acquiring voice data, face image data including a face, and input data including a purpose of the image; determining the movement of the character based on the movement of the facial feature extracted from the facial image data and the voice data; determining a shot corresponding to the character based on the voice data and the purpose; and generating an image corresponding to the audio data based on the determined shot.
- the determining of the shot may include: determining a length of the shot based on an utterance section in the voice data; and determining the type of the shot based on the purpose.
- the type of the shot may be classified by the size of the shot based on the size of the character projected on the shot and the angle of the shot based on the angle of the character projected on the shot.
- the determining of the shot may include, based on the purpose, determining a sequence of a plurality of shots, the plurality of shots including a plurality of shots having different types of shots; dividing the voice data into a plurality of utterance sections based on a change in the size of the voice data; and determining the lengths of the plurality of shots based on the plurality of utterance sections.
- the determining of the lengths of the plurality of shots may include: determining at least one transition point at which a shot is switched based on the purpose and the plurality of utterance sections; and determining the lengths of the plurality of shots based on the turning point.
- the determining of the shot may include: changing an order of shots in the sequence based on a user input; adding at least one shot to the sequence based on a user input; deleting at least one shot in the sequence based on a user input; changing a type of shot in the sequence based on a user's input; and changing the length of the shot in the sequence based on the user's input.
- the determining of the movement of the character may include: determining the movement of a mouth shape of the character based on pronunciation information corresponding to the voice data; and determining the movement of the facial element of the character based on the movement of the facial feature extracted corresponding to the plurality of frames of the facial image data.
- the determining of the movement of the character may include: determining the facial expression of the character based on the purpose; determining a movement of a facial element of the character based on the movement of the facial feature and the voice data; and combining the determined facial expression of the character and the movement of the facial element of the character.
- the determining of the facial expression of the character may further include changing the facial expression of the character based on a user input.
- the obtaining of the input data may further include extracting the movement of the facial feature including at least one of a movement of a pupil, a movement of an eyelid, a movement of an eyebrow, and a movement of a head from the facial image data.
- the character may include: a first character whose movement is determined based on a movement of a first facial feature obtained from first facial image data in the face image data and first voice data in the voice data; and a second character whose movement is determined based on a movement of a second facial feature obtained from second facial image data in the face image data and second voice data in the voice data.
- the determining of the shot includes determining a shot corresponding to the first character and the second character based on the first voice data in the voice data, the second voice data in the voice data, and the purpose can do.
- the determining of the shot may include determining the arrangement of the first character and the second character included in the shot, based on the purpose.
- the determining of the movement of the character may include: determining an interaction between the first character and the second character based on the purpose and at least one of the first voice data and the second voice data; and determining the movement of the first character and the movement of the second character based on the determined interaction.
- the voice data may include first voice data acquired from the first user terminal and second voice data acquired from the second user terminal.
- the face image data may include first face image data obtained from the first user terminal and second face image data obtained from the second user terminal.
- An image generating apparatus obtains voice data, face image data including a face, and input data including a purpose of an image, and based on the movement of facial features extracted from the face image data and the voice data, at least one processor that determines the movement of the character, determines a shot corresponding to the character based on the voice data and the purpose, and generates an image corresponding to the voice data based on the determined shot includes
- the processor determines a sequence of a plurality of shots, the plurality of shots including a plurality of shots having different types of shots, based on the purpose, and a size of the voice data. Based on the change, the voice data may be divided into a plurality of speech sections, and lengths of the plurality of shots may be determined based on the plurality of speech sections.
- the processor may perform an operation of changing an order of shots in the sequence based on a user input; adding at least one shot to the sequence based on a user input; deleting at least one shot in the sequence based on a user's input; changing a type of shot in the sequence based on a user's input; and changing the length of the shot in the sequence based on the user's input.
- the processor determines the expression of the character based on the purpose, and determines the movement of the facial element of the character based on the movement of the facial feature and the voice data, The determined facial expression of the character and the movement of the facial element of the character may be combined.
- the processor may change the facial expression of the character based on a user's input when determining the facial expression of the character.
- FIG. 1 is an operation flowchart of an image generating method according to an exemplary embodiment.
- FIG. 2 is a diagram illustrating an example of various facial expressions corresponding to a character according to an embodiment.
- 3 is a diagram illustrating examples of shots of different types determined according to the purpose of an image, according to an embodiment.
- FIG. 4 is a diagram illustrating an example of a graph of a change in the volume of a sound according to the passage of time according to an embodiment.
- 5 to 9 are diagrams illustrating examples of an interface for editing a determined shot according to an embodiment.
- FIG. 10 is a block diagram illustrating a configuration of an image generating system according to an exemplary embodiment.
- FIG. 1 is an operation flowchart of an image generating method according to an exemplary embodiment.
- an image generating method includes obtaining input data ( 110 ), determining a character ( 120 ), determining a movement of a character ( 130 ), and a shot corresponding to the character. It may include the step of determining 140 , and the step of generating an image corresponding to the audio data ( 150 ).
- the image generating method may be performed by at least one processor of the image generating system.
- the image generating system is a system that processes voice data, face image data including a face, and input data including a purpose of the image, and outputs an image corresponding to the input data, and may be driven by at least one processor.
- the image generating system may be briefly referred to as a system.
- the image generating system may be implemented as a device, and the device in which the image generating system is implemented may include a user terminal (eg, a mobile phone, a computer) and a server.
- the user terminal may include a user interface for receiving input data from a user and providing the user with an image according to the processing of the input data.
- the user terminal may be operatively connected to the server through a network.
- the user terminal may be briefly referred to as a terminal.
- the operations of the image generating method according to an embodiment may be performed by a processor constituting a terminal or a server in the system. A detailed configuration of the image generating system according to an embodiment will be described in detail with reference to FIG. 10 below.
- Step 110 may include acquiring voice data, face image data including a face, and input data including a purpose of the image.
- the voice data may correspond to a sound signal generated by the user's utterance.
- the face image data is image data including at least one face, and may include a still image and a moving image.
- the face image data will be described using a moving picture as an example, but is not limited thereto.
- the face image data and the voice data may correspond to data synchronized with the time axis.
- the acquired face image data may correspond to a moving picture of the uttering user's face
- the acquired voice data may include a uttering user's voice signal synchronized with the face image data on a time axis. .
- the purpose of a video is to relate to the subject or style of the video being created, for example, sincere assertions, strong will, objective analysis, romantic conversations, quarrels, pleasant conversations, fun things, scary conversations, gratitude, sad experiences, It can include embarrassing situations and unfair things.
- the input data may be obtained from a user through a user interface for input.
- the user may input voice data and face image data by shooting a video through a user interface provided by an application installed in the user terminal.
- the user may be provided with predetermined options selectable for the purpose of the image to the user through a user interface provided by an application installed on the user terminal, and at least one selected by the user from among the options regarding the purpose of the predetermined image It can be acquired for the purpose of the acquired image.
- Steps 120 to 140 may include setting parameters for image generation based on the obtained input data.
- the parameter for image generation is a factor for determining at least one element included in the image, and may include, for example, a parameter related to an appearance of a character, a parameter related to a movement of a character, and a parameter related to a shot, but is not limited thereto. it is not
- the input data obtained in step 110 may include a user's selection input for determining the appearance of the character.
- step 120 may include determining a character based on a user's selection input for determining the appearance of the character obtained in step 110 . Determining the character may mean determining the value of a parameter related to the appearance of the character.
- the parameter value for the appearance of the character may include parameter values for each of the elements related to the appearance that determine the appearance of the character (eg, hairstyle, face shape, body type, skin color, eyebrows, eyes, nose, mouth, clothing). can The appearance of the character may be visually determined based on a parameter value related to the appearance of the character.
- parameters related to various appearances of the character may be determined by a user's selection input for determining the appearance of the character.
- a user's selection input for determining the appearance of the character.
- parameter values related to the type of character eg, alien, animal, human
- parameter values related to the character's style eg, Japanese manga style, American comic style
- a plurality of predetermined candidates may be provided to the user in response to elements related to the appearance of the character, and the user selects any one of the plurality of candidates provided in response to each element. It is possible to determine the parameter values of the elements related to the appearance. For example, the user determines parameter values of elements related to the appearance of the character by selecting any one of a plurality of hairstyles, a plurality of face shapes, a plurality of eye shapes, a plurality of skin colors, and a plurality of clothes provided through the interface.
- the plurality of candidates may include elements related to the appearance of a character created by another user.
- operation 120 may include determining a character based on facial features extracted from the facial image data obtained in operation 110 .
- the character may be determined based on the facial features extracted from the facial image data.
- the facial feature is a feature of facial elements recognized in a facial image, for example, a position of a pupil obtained based on a position of a feature point(s) of a face contour, eyes, nose, mouth, and eyebrows, and an eyebrow. It can include the shape of the head and the direction of the head.
- Facial features may be extracted from the face image data based on various face recognition algorithms or facial feature extraction algorithms.
- a value of a parameter related to an appearance of a character may be determined based on a facial feature extracted from facial image data. For example, a parameter value related to the eyebrow element of the character may be determined based on the extracted eyebrow shape, a parameter value related to the character's head element may be determined based on the extracted head direction, and the extracted pupil position may be determined. Based on the parameter values for the eye element of the character may be determined.
- the character may be further determined based on other features in addition to the facial features extracted from the face image data.
- a parameter value related to the appearance of a character may be determined based on characteristics of gender, skin color, face shape, and hairstyle extracted from face image data.
- a parameter value determined based on a facial feature or other features extracted from facial image data may also be changed by a user's selection input. For example, the eyebrow shape, gender, skin color, face shape, and hairstyle of the character already determined by the user's selection input may be changed.
- operation 130 may include determining the movement of the character based on the movement of the facial feature extracted from the facial image data and voice data.
- the movement of the facial feature may include a change in position and shape of at least one facial feature extracted in response to a plurality of frames of facial image data over time. For example, a movement of the pupil including a change in the position of the pupil, a movement of the eyelid including a blinking movement of the eyelid, a movement of the eyebrow including a change in the height and an angle of the eyebrow, and a change in the direction of the head may include movement.
- the movement of the character may be determined based on the movement of the facial feature extracted from the facial image data obtained in step 110 and the voice data obtained in step 110 .
- the movement of the character is a change in the position and/or shape of the entire character or a part of the character over time, and may include a movement corresponding to the movement of a facial feature extracted from facial image data.
- the movement of the character may include movement over time of the position and/or shape of the character itself, the character's eyebrows, pupils, eyelids, arms, legs, or mouth corresponding to the movement of a facial feature. .
- Step 130 includes the steps of determining the movement of the character's mouth shape based on the pronunciation information corresponding to the voice data obtained in step 110 based on the pronunciation information of the voice data ( The method may include determining the movement of the facial element of the character based on the movement of the facial feature extracted in response to the plurality of frames of the facial image data obtained in step 110).
- a mouth shape corresponding to the pronunciation information may be predetermined.
- the first mouth shape corresponding to the pronunciation [a] may be predetermined, and may be determined with reference to the actual mouth shape of a person for making the corresponding pronunciation sound.
- pronunciation information corresponding to each frame of the voice data may be acquired.
- a mouth shape corresponding to the pronunciation information of each frame may be determined based on the mouth shape predetermined in response to the pronunciation information, and by connecting the mouth shape determined corresponding to each frame in the chronological order of the frames, the movement of the mouth shape may be can be created
- the movement of the facial element of the character may be determined based on the movement of the facial feature extracted corresponding to a plurality of frames of the facial image data.
- the movement of the pupil of the character may be determined based on the change in the position of the pupil extracted in response to the plurality of frames, and the head of the character based on the change in the direction of the pupil extracted in response to the plurality of frames. movement can be determined.
- the face element may include facial elements such as eyebrows and eyes included in the character's face except for the shape of the mouth.
- Step 130 may include determining the facial expression of the character based on the obtained purpose, determining the movement of the facial element of the character based on the obtained movement of facial features and voice data, and determining the movement of the character. It may include combining the facial expression and the determined movement of the facial element of the character.
- the face element may include a mouth shape.
- face shapes of a plurality of characters corresponding to a plurality of facial expressions may be predetermined in response to the character determined in step 120 .
- face shapes 201 to 205 of the character corresponding to a plurality of facial expressions corresponding to the determined character may be determined.
- the movement of the mouth shape determined based on pronunciation information corresponding to the voice data may be combined with the shape of the character's face corresponding to the specific expression to generate the movement of the character uttering the specific expression.
- some parameters of the mouth shape may be corrected according to the combined facial expression.
- the parameter of the mouth shape whose movement is determined according to the pronunciation information may be corrected to have a more open mouth or a raised corner of the mouth.
- the parameter of the shape of the mouth whose movement is determined according to the pronunciation information may be corrected in a form in which the mouth is wider to the side.
- the movement of the character corresponding to the obtained facial image data may be generated by combining the movement of the pupil determined based on the movement of the extracted facial feature to the shape of the face of the character corresponding to the specific expression.
- the facial expression of the character determined based on the purpose of the image may include a plurality of facial expressions.
- the facial expression of the character may include a facial expression sequence including a plurality of facial expressions.
- the expression of the character may be determined as the first expression from the first frame to the second frame of the face image data or the voice data, and the second expression from the third frame to the fourth frame.
- step 140 may correspond to a step of determining a shot corresponding to a character based on the purpose of the audio data and image obtained in step 110 .
- a shot is a basic unit of shooting, meaning a scene shot at once, and the type of shot size based on the size of the projected subject (e.g. close-up shot, bust shot, full shot, long shot), Shot types may be classified into types (eg, bird-eye view shots, high-angle shots, eye-level shots, and low-angle shots) based on the angle of the projected subject.
- the types of shots include types based on the number of subjects included in the shot (eg one-shot, two-shot, three-shot, group shot), types based on camera manipulation (eg panning shot, tilt shot, zoom shot), It can include different types, such as types based on the movement of the camera (eg Dali Shots, Tracking Shots, Arc Shots, Aerial Shots).
- the subject may correspond to a character whose movement is determined in steps 120 to 130
- the determined shot may correspond to a scene in which a character moving with a virtual camera is captured in a virtual space. have.
- Step 140 may include determining a length of a shot based on an utterance section in the acquired voice data and determining a type of a shot based on the acquired purpose.
- the determining of the shot 140 includes determining a sequence of a plurality of shots based on the purpose of the image, and dividing the voice data into a plurality of utterance sections based on a change in the size of the voice data.
- the method may include classifying and determining lengths of the plurality of shots based on the plurality of utterance sections.
- Step 140 may include determining a shot sequence including at least one shot corresponding to a purpose.
- the shot sequence may include one shot or a plurality of shots in which an order is determined.
- the plurality of shots included in the shot sequence may include a plurality of shots having different types of shots.
- the plurality of shots included in the shot sequence may include a plurality of shots in which at least one of a shot size and a shot angle is different from each other.
- a shot sequence corresponding to a purpose may be determined based on a shot used when capturing a captured image for the corresponding purpose. For example, referring to FIG. 3 , if the purpose of the image is 'interview', a shot sequence including four shots used when shooting an image for the purpose of 'interview' may be determined corresponding to the purpose of 'interview'. .
- the plurality of shots included in the shot sequence may include different types of shots.
- the plurality of shots included in the shot sequence may include first to fourth type shots in which at least one of a size and an angle of a shot is different from each other. For example, the shot 301 and the shot 302 are shooting the subject up to the chest.
- the size of the shot is the same, but the shot 301 is the front, and the shot 302 is the oblique direction.
- the angle of the shot is different.
- the angle of the bar shot for shooting the subject from the front is the same, but the shot 303 is taken to the neck of the subject, and the shot 301 is taken to the chest of the subject.
- the size of the bar shot you are doing is different.
- Lengths of a plurality of shots included in a shot sequence may be determined based on an utterance period in voice data.
- An utterance section is a section in which utterances are continued in voice data, for example, a section from a point recognized as a starting point of utterance in voice data to a point recognized as an end point of the utterance or a section recognized as a starting point of utterance in voice data.
- a section from a point to a point recognized as a start point of the next utterance of the corresponding utterance may correspond to one utterance section.
- a start point of an utterance or an end point of an utterance may be recognized from the voice data based on a change in the volume over time included in the acquired voice data.
- FIG. 4 shows a graph of a change in the volume of a sound over time corresponding to the acquired voice data.
- a start point or an end point of an utterance may be recognized based on a change in the volume of a sound, and the speech data includes utterance sections 411 to 415 bordering on the recognized start or end point of the utterance.
- the starting point or ending point of the utterance may be determined based on a point at which the loudness of the sound signal changes abruptly, a point at which the volume of the sound signal decreases below a threshold value, and a point at which the level of the sound signal increases above the threshold value, but It is not limited.
- a point at which a sentence starts or a point at which a sentence ends may be determined as a start point or an end point of an utterance by using the speech recognition result of the voice data.
- the start point of the specific utterance may be the same as the end point of the previous utterance, or may be recognized later than the end point of the previous utterance. For example, if the time interval between a point recognized as the start point of a specific utterance and a point recognized as the end point of the previous utterance is less than a predetermined threshold, the start point of the specific utterance and the end point of the previous utterance are the same point. can be decided. For example, referring to FIG. 4 , the end point 403 of the utterance corresponding to the section 412 and the start point 403 of the utterance corresponding to the section 413 may be determined as the same point.
- the end point 404 of the utterance corresponding to the section 413 and the start point 405 of the utterance corresponding to the section 415 may be determined as different points, and in this case, the section 414 is the speech signal of the utterance. may not be included, but the section 414 may also be included in the utterance section.
- the length of the shot may be determined to include at least one utterance section.
- the first shot in a temporally advanced order starts from a start point 401 of voice data, a point 402, a point 403,
- the length to either point 404 or point 405 may be determined.
- the first shot may be determined to have a length including any one of 1 to 4 sections from the start point 401 of the voice data.
- the second shot that is the next shot of the first shot may be determined to have a length including any one of 1 to 4 sections from the end point of the first shot to the end point 406 of the voice data.
- the determining of the lengths of the plurality of shots includes determining at least one turning point at which the shot is switched based on the purpose of the image and the plurality of utterance sections, and based on the determined turning point, determining the lengths of the shots of A boundary of an utterance section included in the voice data may be determined as a transition point at which a shot transition occurs.
- the determined number of transition points may correspond to the number of shots included in the shot sequence.
- the chronological first shot included in the shot sequence may be determined as the length from the start point of the voice data to the first turning point in chronological order
- the second shot may be determined as the length from the first turning point to the second turning point.
- the last shot may be determined as the length from the end point of the previous shot to the end point of the voice data.
- the number of shots included in the shot sequence and the number of transition points determined from the voice data may be determined based on the purpose of the acquired image. For example, when the purpose of the video is to express a sense of excitement, the number of shots included in the shot sequence and the number of transition points in the voice data may be determined so that a large number of shot changes occur.
- the length between shots may be arbitrarily determined or may be determined based on the purpose of the image.
- a transition point in the audio data may be determined so that the lengths of a plurality of shots included in the shot sequence are similarly determined according to the purpose of the image.
- the transition point in the voice data may be determined as a boundary between speech sections included in the voice data.
- a transition point in the audio data may be determined such that the number of audio sections included in each of a plurality of shots included in the shot sequence is the same according to the purpose of the video.
- a ratio of lengths between a plurality of shots included in a shot sequence may be determined according to the purpose of an image, and a transition point in the voice data may be determined to follow the determined ratio.
- the posture and posture movement of the character included in the shot may be determined based on the purpose of the acquired image.
- the posture of the character may be determined as a sitting posture in response to the purpose of the image of the interview, and the movement of the posture may be determined to include a gesture of a hand motion according to an utterance in the sitting posture.
- the background and props included in the shot may be determined based on the purpose of the acquired image.
- the background corresponding to the interview space may be determined in response to the purpose of the video of the interview, and a chair or a camera may be included in the shot as a prop.
- shots corresponding to the purpose of the image may be determined based on a rule-based model. For example, based on a predefined rule, a set of shots corresponding to a specific purpose or a sequence of shots corresponding to a specific purpose may be determined. When a set of shots corresponding to a specific purpose is determined based on a predefined rule, an order between the shots may be determined based on the predefined rule. The order between the shots may be arbitrarily determined or may be determined based on a predetermined priority between the shots.
- shots corresponding to the purpose of the image may be determined based on a statistical model.
- a statistics-based model is a set or shot of shots corresponding to an image of a specific purpose based on statistical probability, such as the frequency of shots appearing in an image for a specific purpose, and a probability that a second shot occurs after the first shot in an image of a specific purpose. It may correspond to a model that outputs a sequence of
- shots corresponding to the purpose of the image may be determined based on a learning-based model.
- the learning-based model learns the purpose of the image and the type of shot used in the image data from the image data, and when the purpose of the image is input, it may correspond to a model that outputs a set of shots or a sequence of shots corresponding to the input purpose.
- the learning-based model may include a deep learning model, and the structure of the deep learning model may be configured in various ways. According to the structure of the learning-based model, it may be learned based on various learning methods.
- the order between the shots may be arbitrarily determined or may be determined based on the second model.
- a method of determining a set of shots corresponding to a specific purpose and a method of determining an order of shots may use different models. For example, a set of shots corresponding to a specific purpose may be determined using a first model based on statistics, and an order of shots may be determined using a second model based on a rule.
- the lengths of shots included in the shot sequence may be determined based on at least one of a rule-based model, a statistics-based model, and a learning-based model.
- the lengths of the shots included in the shot sequence may be determined based on voice data in a model determining the shot sequence, or may be determined based on the shot sequence and voice data in a model configured independently of the model determining the shot sequence.
- elements included in the shot corresponding to the purpose of the image may be determined based on at least one of a rule-based model, a statistics-based model, and a learning-based model.
- the elements included in the shot may include a posture or posture movement of a character included in the shot, a background or props included in the shot.
- the determined shot and elements included in the shot may be changed based on a user input.
- the order of the shot sequence determined based on the user input may be changed, the type of shot included in the shot sequence may be changed, the lengths of the determined shots may be changed, and the shot included in the shot sequence may be changed. may be added or deleted.
- the appearance, facial expression, and posture of the character included in the shot may be changed based on the user input, and the background and props included in the shot may be changed.
- step 150 may include generating an image corresponding to voice data based on the determined shot.
- an image including a character whose movement is determined may be generated based on the determined shot.
- the generated image may correspond to an image in which a character moving according to the determined movement is sequentially photographed with a length determined by a shot type according to the determined shot sequence.
- the generated image may be output together with input audio data, and may correspond to an image synchronized with the audio data in a time axis.
- the generated image may include a character whose movement is determined to make an utterance corresponding to the voice data.
- a pre-processing operation such as removing noise other than a voice may be performed on the input voice data, and a modulation operation for changing a voice may be performed.
- An image generating method may include generating an image including a plurality of characters.
- generating an image including a plurality of characters For convenience of explanation, a case where the plural number is 2 will be described as an example, but the operation for generating an image including two characters may be applied to an operation for generating an image including two or more characters.
- the plurality of characters includes a first character whose movement is determined based on a movement of a first facial feature obtained from the first face image data and first voice data and a second face obtained from the second face image data. It may include a second character whose movement is determined based on the movement of the feature and the second voice data.
- the first face image data and the first voice data may correspond to input data obtained from the first user terminal
- the second face image data and the second voice data may correspond to input data obtained from the second user terminal. can do.
- the appearance of the first character may be determined based on a selection input for determining the appearance of the character obtained from the first user terminal, and the appearance of the second character may be determined from the appearance of the character obtained from the second user terminal.
- the determination may be based on a selection input regarding the determination.
- the input data received by the image generating system may include real-time video call data of the first user terminal and the second user terminal.
- the first face image data captured by the first user terminal and the first voice data that is the user's voice received through the first user terminal may be received through the first user terminal, and through the second user terminal Second face image data captured by the second user terminal and second voice data that is a user's voice received through the second user terminal may be received.
- the input data received by the image generating system may include first input data received from the first user terminal and second input data received from the second user terminal in relation to the first input data.
- first input data received from the first user terminal
- second input data corresponding to the first image is received from the terminal as the second user.
- a second image following the first image may be generated based on the second input data.
- the second image may be determined based on the second input data, the movement of the first character included in the first image, and other factors included in the first image. For example, if the first image includes a first character sitting on a bench and talking, the second image may be generated to include the second character sitting next to the first character and talking.
- the determining of the movement of the character 130 may include determining the interaction between the first character and the second character based on at least one of the purpose of the image, the first voice data, and the second voice data. and determining the movement of the first character and the movement of the second character based on the determined interaction.
- the interaction may refer to an action including interaction or mutual contact of a plurality of characters, such as talking, shaking hands, or hugging.
- the interaction between the first character and the second character may be determined based on the purpose of the acquired image.
- a plurality of interaction candidates may be predetermined in correspondence with each purpose of the image.
- interactions such as holding hands and hugging may be determined as interaction candidates
- an interaction such as shaking hands may be determined as interaction candidates. .
- an interaction between the first character and the second character may be determined based on acquired voice data.
- An interaction corresponding to a specific word may be determined in advance, and when a corresponding word is recognized in voice data, an interaction corresponding thereto may be determined. For example, when the word 'I love you' is recognized in the acquired first voice data or the second voice data, an interaction of hugging may be determined.
- an interaction corresponding to a recognized word may be determined differently according to the purpose of the acquired image. For example, for a first purpose, when a first word is recognized, a first interaction may be determined, but for a second purpose, when the same first word is recognized, a second interaction different from the first interaction may be determined. According to an embodiment, any one of a plurality of predetermined interaction candidates may be selected in response to the purpose of the image based on the word recognized in the voice data.
- the step of determining the shot 140 may include determining the shot corresponding to the first character and the second character based on the purpose of the first audio data, the second audio data, and the image.
- the shot corresponding to the first character and the second character may include a shot including at least one of the first character and the second character.
- the shot corresponding to the first character and the second character may be set such that only the first character appears, only the second character appears, or both the first character and the second character appear.
- a shot in which the first character is photographed from the front may be determined when the first character speaks, and the first character is photographed from the front when the second character speaks.
- a shot can be determined.
- a shot taken to include the first character and the second character may be determined.
- the determining of the shot 140 may include determining the arrangement of the first character and the second character included in the shot, based on the purpose of the image.
- the arrangement of the first character and the second character may include a positional relationship between the first character and the second character displayed in the shot and a composition of the first character and the second character. For example, based on the purpose of the image, the arrangement may be determined so that the first character and the second character sit facing each other, the arrangement may be determined so that they stand side by side, the first character is standing, and the second character is the second character.
- the arrangement may be determined so that the first character is seated while looking at the first character, and the arrangement in which the first character approaches the second character from behind may be determined.
- the relationship between the first character and the second character may be determined based on the purpose of the image, and the arrangement of the first character and the second character may be determined based on the determined relationship. For example, when the relationship between the first character and the second character is determined to be a friend or a lover, the first character and the second character may be arranged side by side.
- the type of communication or the type of situation may be determined based on the purpose of the image, and the arrangement of the first character and the second character may be determined based on the determined type of communication.
- the first character and the second character may be arranged to face each other.
- the arrangement of the first character and the second character may be determined so as to face the same direction.
- the emotion between the first character and the second character may be determined based on the purpose of the image, and the arrangement of the first character and the second character may be determined based on the determined type of emotion. For example, when the type of emotion is determined as love or gratitude, the first character and the second character may be arranged to face each other, and when the type of emotion is determined as surprise or fear, the first character moves from behind the second character The arrangement can be determined to access.
- the arrangement between the first character and the second character may be determined based on the purpose of the acquired image.
- a plurality of placement candidates may be predetermined in correspondence with each purpose of the image.
- placement candidates such as a side-by-side arrangement on a bench, a face-to-face arrangement, and a hand-held arrangement may be determined.
- the arrangement of the first character and the second character may be determined based on a user input.
- the user may input a request to determine a placement between the first character and the second character through the interface. For example, by selecting any one of predetermined placement candidates between the first character and the second character, a request for determining the placement between the first character and the second character may be input. Placement candidates between the first character and the second character may be predetermined based on the purpose of the image as described above.
- the operations of steps 120 to 140 may be performed in parallel or sequentially.
- the operation 120 of determining the character and the operation 130 of determining the movement of the character may be performed in parallel, and the determined character may be synthesized with the determined movement of the character, and the moving character may be included in the image.
- the operation 140 of determining the shot corresponding to the character may be performed in parallel with the operation 120 of determining the character and the operation 130 of determining the movement of the character, and the operation of determining the shot
- a sequence of shots for taking an arbitrary character at a specific angle and a specific size is determined by 140 , and the moving character determined by operations 120 and 130 is converted into a sequence of shots determined by operation 140 .
- An image may be generated by photographing.
- 5 to 9 show examples of an interface for editing a determined shot according to an embodiment.
- the determined shot sequence may include a plurality of shots, and the plurality of shots may include shots having different types of shots.
- the first shot 501 corresponds to the first type of shot in which the character is photographed from the frontal angle to the chest position
- the second shot 502 is the second shot in which the character is photographed from the frontal angle to the knee position. It can correspond to two types of shots.
- the determined shot sequence may include a plurality of shots of the same type.
- the first shot 501 and the third shot 503 included in the determined shot sequence may correspond to the same type of shot.
- the length of the shot may be determined based on an utterance section in voice data.
- the points 511 to 514 at which the shots are switched are points recognized as the start point or the end point of the speech, and may correspond to the boundary of the speech section.
- the interface may provide an interfacing object 530 for changing a sequence of a plurality of shots determined in response to the purpose of an image, and the user may provide a sequence of shots determined based on the interfacing object 530 .
- the interfacing object 530 may correspond to an interfacing object configured to transmit a command for changing a shot sequence in response to a user input. For example, in response to a shot change input of clicking the interfacing object 530 , the order of shots included in the shot sequence may be changed, the types of shots included in the shot sequence may be changed, and the shot sequence may be changed to the shot sequence. The number of included shots may be changed, and the length of the shots may be changed.
- the shot sequence 510 shown in FIG. 5 is changed to the shot sequence 610 shown in FIG. 6 .
- the number of shots included in the shot sequence 610 may be changed according to a shot change input, and types of shots included in the shot sequence 610 may be changed.
- the change of the shot sequence by the interfacing object 530 may be performed according to a predefined rule.
- a shot sequence may be determined by selecting a part from a set of shots determined according to the purpose of the image according to a predetermined order.
- candidates of a plurality of shot sequences may be output from a model for determining a shot sequence, and may be determined according to a predetermined order from the candidates of the output shot sequences.
- the interface may provide a set 520 of a plurality of shots determined to correspond to the purpose of the image, and the user may change the configuration of the shot sequence determined based on the set 520 of the provided shots.
- the user may input a request to change the shot 505 included in the shot sequence by selecting the shot 521 from the set of shots 520 through the interface, and in response to the user's change input
- the shot sequence 510 may be changed to the shot sequence 710 shown in FIG. 7 .
- the user may input a request for selecting at least one shot from the set of shots 520 and adding it to a specific position of the shot sequence through the interface, and in response to the user's additional input, the shot sequence 510 . ) may be changed to the shot sequence 810 shown in FIG. 8 .
- the user may input a request to change the size or angle of at least some shots included in the determined shot sequence through the interface.
- the user selects at least one shot included in the shot sequence to control the size of the character included in the shot, or controls the angle of the character included in the shot, thereby making a request to change the size or angle of the shot. can be entered.
- a specific type of shot determined based on a user input may be included in the shot sequence.
- the user may input a request to delete at least some shots included in the determined shot sequence through the interface. For example, the user may select a shot included in the shot sequence and request deletion from the shot sequence, and the shot selected by the user may be deleted from the shot sequence. As some shots included in the shot sequence are deleted, the length of a shot positioned before or after the deleted shot may be adjusted.
- the interface may provide a function for adjusting the lengths of the plurality of shots determined in response to the purpose of the image.
- the user may input a request for changing the determined length of the shot sequence by changing the position of the boundary of the shot through the interface.
- the position of the shot boundary may be changed to the boundary of the firing section, and when the position of the shot boundary input by the user does not correspond to the boundary of the firing section, the position of the shot boundary is changed to the boundary of the nearby ignition section by the magnetic effect can be
- the user may input a request to change the position 512 of the boundary of the shot 502 and the shot 503 to the position 515 through the interface, and in response to the user's change input, the shot sequence ( 510 may be changed to the shot sequence 910 shown in FIG. 9 .
- the expression of the character determined based on the purpose of the image or the sequence of the plurality of facial expressions of the character corresponding to the image may be changed by a user's selection input.
- the user may change the facial expression determined in response to each frame to another facial expression by a selection input for a plurality of facial expressions provided through the interface in response to the character, and by adding another facial expression to the facial expression sequence, each frame It is possible to determine the expression corresponding to the
- visual elements and auditory elements corresponding to a specific shot may be determined by a user's input.
- visual elements such as postures, props, backgrounds, and texts of a character corresponding to the entire shot or some shots may be set or changed by a user input.
- auditory elements such as sound effects corresponding to a specific shot and background music corresponding to all shots or some shots may be set or changed according to a user input.
- elements set or changed by a user's input may be determined by an input of selecting some of a plurality of candidates provided through an interface.
- a plurality of candidates regarding various postures of the character may be provided through the interface, and the user selects any one of a plurality of candidates regarding the posture of the character, thereby determining the posture of the character in the scene determined by the shot or The sequence of postures can be determined.
- a plurality of candidates for sound effect may be provided through the interface, and the user may determine the sound effect by selecting the sound effect to be included in the image at a specific time from among the plurality of candidates.
- FIG. 10 is a block diagram illustrating a configuration of an image generating system according to an exemplary embodiment.
- an image generating system may be implemented as a device, and the device in which the image generating system is implemented may include a terminal 1010 and a server 1020 .
- the terminal 1010 and the server 1020 may correspond to an image generating apparatus that performs the above-described image generating method.
- the configuration of the system shown in FIG. 10 is an example for description of the invention, and the number of terminals or the number of servers included in the system is not limited as in FIG. 10 .
- the terminal 1010 may be a fixed terminal implemented as a computer device or a mobile terminal.
- the terminal 1010 may include a smart phone, a mobile phone, a computer, a notebook computer, a digital broadcasting terminal, a personal digital assistant (PDA), a portable multimedia player (PMP), and a tablet PC.
- PDA personal digital assistant
- PMP portable multimedia player
- the user terminal may communicate with the server 1020 and/or other electronic devices through the network 1030 using a wireless or wired communication method.
- the server 1020 may be implemented as a computer device or a plurality of computer devices that communicate with the terminal 1010 and the network 1030 to provide commands, codes, files, contents, services, and the like.
- the communication method is not limited, and may include a communication method using a communication network (eg, a mobile communication network, a wired Internet, a wireless Internet, a broadcasting network) that the network 1030 may include, and a short-range wireless communication method between devices.
- the network 1030 may include a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), and a broadband network (BBN). , the Internet, and the like.
- PAN personal area network
- LAN local area network
- CAN campus area network
- MAN metropolitan area network
- WAN wide area network
- BBN broadband network
- the server 1020 may provide a file for installing an application to the terminal 1010 accessed through the network 1030 .
- the terminal 1010 may install an application using a file provided from the server 1020 .
- the service or content provided by the server 1020 by accessing the server 1020 under the control of an operating system (OS) and at least one program (eg, a browser or an installed application) included in the terminal 1010 . can be provided.
- OS operating system
- the server 1020 transmits a code corresponding to the service request message to the terminal 1010 .
- the terminal 1010 can provide content to the user by composing and displaying a screen according to the code according to the control of the application.
- the terminal 1010 and the server 1020 may include memories 1011 and 1021 , processors 1013 and 1023 , communication modules 1015 and 1025 , and input/output interfaces 1017 and 1027 . .
- the memories 1011 and 1021 may store information related to the image generating method described above with reference to FIGS. 1 to 9 or a program implementing the image generating method described above.
- the memories 1011 and 1021 are computer-readable recording media, and may be volatile memory or non-volatile memory.
- the program in which the above-described image generating method is implemented may include a code for a browser or application installed and driven in the terminal 1010 by files provided from the server 1020 through the network 1030 . .
- the communication modules 1015 and 1025 may provide a function for the terminal 1010 and the server 1020 to communicate with each other through the network 1030, and for communication with other electronic devices or other servers. function can be provided. For example, a request generated by the processor 1013 of the terminal 1010 according to a program code stored in a recording device such as the memory 1011 is controlled by the communication module 1015 through the network 1030 to the server 1020 can be transmitted to For example, a control signal, command, content, file, etc. provided under the control of the processor 1023 of the server 1020 passes through the communication module 1025 and the network 1030 to the communication module 1015 of the terminal 1010 .
- a control signal or command of the server 1020 received through the communication module 1015 may be transmitted to the processor 1013 or the memory 1011 , and the terminal 1010 further includes content or files. It may be stored as a storage medium that can be
- the input/output interfaces 1017 and 1027 may be means for interfacing with the input/output device 1019 .
- the input device may include a device such as a keyboard or mouse
- the output device may include a device such as a display for displaying a communication session of an application.
- the input/output interface 1017 may be a means for an interface with a device in which functions for input and output are integrated into one, such as a touch screen.
- the processor 1013 of the terminal 1010 processes the command of the computer program loaded in the memory 1011, and the service screen or content configured using the data provided by the server 1020 is displayed on the input/output interface ( 1017) can be displayed on the display.
- An input received from the user through the input/output device 1019 may be provided in a form that can be processed by the processor 1013 of the terminal 1010 through the input/output interface 1017 .
- the terminal 1010 and the server 1020 may include other components not shown in FIG. 10 .
- the terminal 1010 is implemented to include at least a portion of the above-described input/output device 1019 or other components such as a transceiver, a global positioning system (GPS) module, a camera, various sensors, and a database. It may include more.
- GPS global positioning system
- the embodiments described above may be implemented by a hardware component, a software component, and/or a combination of a hardware component and a software component.
- the apparatus, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using a general purpose computer or special purpose computer.
- the processing device may execute an operating system (OS) and a software application running on the operating system.
- the processing device may also access, store, manipulate, process, and generate data in response to execution of the software.
- OS operating system
- the processing device may also access, store, manipulate, process, and generate data in response to execution of the software.
- the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.
- the method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium.
- the computer readable medium may store program instructions, data files, data structures, etc. alone or in combination, and the program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software.
- Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks.
- - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like.
- Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.
- the hardware devices described above may be configured to operate as one or a plurality of software modules to perform the operations of the embodiments, and vice versa.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Signal Processing (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
Abstract
Description
Claims (20)
- 음성 데이터, 얼굴을 포함하는 얼굴 영상 데이터 및 영상의 목적을 포함하는 입력 데이터를 획득하는 단계;상기 얼굴 영상 데이터로부터 추출된 얼굴 특징의 움직임 및 상기 음성 데이터에 기초하여, 캐릭터의 움직임을 결정하는 단계;상기 음성 데이터 및 상기 목적에 기초하여, 상기 캐릭터에 대응하는 샷을 결정하는 단계; 및상기 결정된 샷에 기초하여, 상기 음성 데이터에 대응하는 영상을 생성하는 단계를 포함하는,영상 생성 방법.
- 제1항에 있어서,상기 샷을 결정하는 단계는상기 음성 데이터 내 발화 구간에 기초하여, 상기 샷의 길이를 결정하는 단계; 및상기 목적에 기초하여, 상기 샷의 유형을 결정하는 단계를 포함하는,영상 생성 방법.
- 제2항에 있어서,상기 샷의 유형은상기 샷에 투영된 캐릭터의 크기에 기초한 샷의 크기 및 상기 샷에 투영된 캐릭터의 각도에 기초한 샷의 앵글에 의해 구분되는,영상 생성 방법.
- 제1항에 있어서,상기 샷을 결정하는 단계는상기 목적에 기초하여, 복수의 샷들- 상기 복수의 샷들은 샷의 유형이 서로 다른 복수의 샷들을 포함함 -의 시퀀스를 결정하는 단계;상기 음성 데이터의 크기 변화에 기초하여, 상기 음성 데이터를 복수의 발화 구간들로 구분하는 단계; 및상기 복수의 발화 구간들에 기초하여, 상기 복수의 샷들의 길이들을 결정하는 단계를 포함하는,영상 생성 방법.
- 제4항에 있어서,상기 복수의 샷들의 길이들을 결정하는 단계는상기 목적 및 상기 복수의 발화 구간들에 기초하여, 샷이 전환되는 적어도 하나의 전환점을 결정하는 단계; 및상기 전환점에 기초하여, 상기 복수의 샷들의 길이들을 결정하는 단계를 포함하는,영상 생성 방법.
- 제4항에 있어서,상기 샷을 결정하는 단계는사용자의 입력에 기초하여, 상기 시퀀스 내 샷들의 순서를 변경하는 단계;사용자의 입력에 기초하여, 상기 시퀀스에 적어도 하나의 샷을 추가하는 단계;사용자의 입력에 기초하여, 상기 시퀀스 내 적어도 하나의 샷을 삭제하는 단계;사용자의 입력에 기초하여, 상기 시퀀스 내 샷의 유형을 변경하는 단계; 및사용자의 입력에 기초하여, 상기 시퀀스 내 샷의 길이를 변경하는 단계중 적어도 하나를 더 포함하는,영상 생성 방법.
- 제1항에 있어서,상기 캐릭터의 움직임을 결정하는 단계는상기 음성 데이터에 대응하는 발음 정보에 기초하여, 상기 캐릭터의 입 모양의 움직임을 결정하는 단계; 및상기 얼굴 영상 데이터의 복수의 프레임들에 대응하여 추출된 상기 얼굴 특징의 움직임에 기초하여, 상기 캐릭터의 얼굴 요소의 움직임을 결정하는 단계를 포함하는,영상 생성 방법.
- 제1항에 있어서,상기 캐릭터의 움직임을 결정하는 단계는상기 목적에 기초하여 상기 캐릭터의 표정을 결정하는 단계;상기 얼굴 특징의 움직임 및 상기 음성 데이터에 기초하여 상기 캐릭터의 얼굴 요소의 움직임을 결정하는 단계; 및상기 결정된 캐릭터의 표정 및 상기 캐릭터의 얼굴 요소의 움직임을 결합하는 단계를 포함하는,영상 생성 방법.
- 제8항에 있어서,상기 캐릭터의 표정을 결정하는 단계는사용자의 입력에 기초하여, 상기 캐릭터의 얼굴 표정을 변경하는 단계를 더 포함하는,영상 생성 방법.
- 제1항에 있어서,상기 입력 데이터를 획득하는 단계는상기 얼굴 영상 데이터에서 눈동자의 움직임, 눈꺼풀의 움직임, 눈썹의 움직임 및 고개의 움직임 중 적어도 하나를 포함하는 상기 얼굴 특징의 움직임을 추출하는 단계를 더 포함하는,영상 생성 방법.
- 제1항에 있어서,상기 캐릭터는상기 얼굴 영상 데이터 내 제1 얼굴 영상 데이터로부터 획득된 제1 얼굴 특징의 움직임 및 상기 음성 데이터 내 제1 음성 데이터에 기초하여 움직임이 결정되는 제1 캐릭터; 및상기 얼굴 영상 데이터 내 제2 얼굴 영상 데이터로부터 획득된 제2 얼굴 특징의 움직임 및 상기 음성 데이터 내 제2 음성 데이터에 기초하여 움직임이 결정된 제2 캐릭터를 포함하고,상기 샷을 결정하는 단계는상기 음성 데이터 내 제1 음성 데이터, 상기 음성 데이터 내 제2 음성 데이터 및 상기 목적에 기초하여, 상기 제1 캐릭터 및 상기 제2 캐릭터에 대응하는 샷을 결정하는 단계를 포함하는,영상 생성 방법.
- 제11항에 있어서,상기 샷을 결정하는 단계는상기 목적에 기초하여, 상기 샷에 포함된 상기 제1 캐릭터 및 상기 제2 캐릭터의 배치를 결정하는 단계를 포함하는,영상 생성 방법.
- 제11항에 있어서,상기 캐릭터의 움직임을 결정하는 단계는상기 목적, 상기 제1 음성 데이터 및 상기 제2 음성 데이터 중 적어도 하나에 기초하여, 상기 제1 캐릭터 및 상기 제2 캐릭터의 인터랙션을 결정하는 단계; 및상기 결정된 인터랙션에 기초하여, 상기 제1 캐릭터의 움직임 및 상기 제2 캐릭터의 움직임을 결정하는 단계를 더 포함하는,영상 생성 방법.
- 제11항에 있어서,상기 음성 데이터는 제1 사용자 단말로부터 획득된 제1 음성 데이터 및 제2 사용자 단말로부터 획득된 제2 음성 데이터를 포함하고,상기 얼굴 영상 데이터는 상기 제1 사용자 단말로부터 획득된 제1 얼굴 영상 데이터 및 상기 제2 사용자 단말로부터 획득된 제2 얼굴 영상 데이터를 포함하는,영상 생성 방법.
- 하드웨어와 결합되어 제1항의 방법을 실행시키기 위하여 매체에 저장된 컴퓨터 프로그램.
- 음성 데이터, 얼굴을 포함하는 얼굴 영상 데이터 및 영상의 목적을 포함하는 입력 데이터를 획득하고,상기 얼굴 영상 데이터로부터 추출된 얼굴 특징의 움직임 및 상기 음성 데이터에 기초하여, 상기 캐릭터의 움직임을 결정하고,상기 음성 데이터 및 상기 목적에 기초하여, 캐릭터에 대응하는 샷을 결정하며,상기 결정된 샷에 기초하여, 상기 음성 데이터에 대응하는 영상을 생성하는,적어도 하나의 프로세서를 포함하는,영상 생성 장치.
- 제16항에 있어서,상기 프로세서는,상기 샷을 결정함에 있어서,상기 목적에 기초하여, 복수의 샷들- 상기 복수의 샷들은 샷의 유형이 서로 다른 복수의 샷들을 포함함 -의 시퀀스를 결정하고,상기 음성 데이터의 크기 변화에 기초하여, 상기 음성 데이터를 복수의 발화 구간들로 구분하고,상기 복수의 발화 구간들에 기초하여, 상기 복수의 샷들의 길이들을 결정하는,영상 생성 장치.
- 제17항에 있어서,상기 프로세서는,상기 샷을 결정함에 있어서,사용자의 입력에 기초하여, 상기 시퀀스 내 샷들의 순서를 변경하는 동작;사용자의 입력에 기초하여, 상기 시퀀스에 적어도 하나의 샷을 추가하는 동작;사용자의 입력에 기초하여, 상기 시퀀스 내 적어도 하나의 샷을 삭제하는 동작;사용자의 입력에 기초하여, 상기 시퀀스 내 샷의 유형을 변경하 동작; 및사용자의 입력에 기초하여, 상기 시퀀스 내 샷의 길이를 변경하는 동작중 적어도 하나를 더 수행하는,영상 생성 장치.
- 제16항에 있어서,상기 프로세서는,상기 캐릭터의 움직임을 결정함에 있어서,상기 목적에 기초하여 상기 캐릭터의 표정을 결정하고,상기 얼굴 특징의 움직임 및 상기 음성 데이터에 기초하여 상기 캐릭터의 얼굴 요소의 움직임을 결정하며,상기 결정된 캐릭터의 표정 및 상기 캐릭터의 얼굴 요소의 움직임을 결합하는,영상 생성 장치.
- 제19항에 있어서,상기 프로세서는,상기 캐릭터의 표정을 결정함에 있어서,사용자의 입력에 기초하여, 상기 캐릭터의 얼굴 표정을 변경하는,영상 생성 장치.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2023562502A JP2024513515A (ja) | 2021-04-08 | 2021-11-26 | 映像生成方法及び装置 |
CN202180099111.0A CN117441192A (zh) | 2021-04-08 | 2021-11-26 | 图像生成方法和装置 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020210045651A KR102345729B1 (ko) | 2021-04-08 | 2021-04-08 | 영상 생성 방법 및 장치 |
KR10-2021-0045651 | 2021-04-08 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022215823A1 true WO2022215823A1 (ko) | 2022-10-13 |
Family
ID=79348226
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2021/017653 WO2022215823A1 (ko) | 2021-04-08 | 2021-11-26 | 영상 생성 방법 및 장치 |
Country Status (6)
Country | Link |
---|---|
US (1) | US20220328070A1 (ko) |
EP (1) | EP4071760A1 (ko) |
JP (1) | JP2024513515A (ko) |
KR (2) | KR102345729B1 (ko) |
CN (1) | CN117441192A (ko) |
WO (1) | WO2022215823A1 (ko) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102601159B1 (ko) * | 2022-09-30 | 2023-11-13 | 주식회사 아리아스튜디오 | 버추얼 휴먼 인터렉션 생성 장치 및 그에 대한 방법 |
US12023160B1 (en) * | 2023-06-16 | 2024-07-02 | Carlos Andrés Cuestas Rodríguez | Non-invasive remote system and method to determine the probability of deceit based on artificial intelligence |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140218373A1 (en) * | 1999-05-21 | 2014-08-07 | Quonsil Pl. 3, Llc | Script control for camera positioning in a scene generated by a computer rendering engine |
KR20190046371A (ko) * | 2017-10-26 | 2019-05-07 | 에스케이텔레콤 주식회사 | 얼굴 표정 생성 장치 및 방법 |
KR20190087230A (ko) * | 2018-01-16 | 2019-07-24 | 삼성전자주식회사 | 복수의 카메라를 이용하여 영상 데이터를 생성하는 방법 및 서버 |
KR20200017266A (ko) * | 2018-08-08 | 2020-02-18 | 삼성전자주식회사 | 아바타에 대응하는 속성에 따른 아이템을 제공하는 방법 및 장치 |
KR20200046417A (ko) * | 2018-10-24 | 2020-05-07 | 삼성전자주식회사 | 보정 이미지 기반 아바타 생성 방법 및 장치 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5326910B2 (ja) * | 2009-01-20 | 2013-10-30 | ソニー株式会社 | 情報処理装置、情報処理方法およびプログラム |
US8726161B2 (en) * | 2010-10-19 | 2014-05-13 | Apple Inc. | Visual presentation composition |
US10290320B2 (en) * | 2015-12-09 | 2019-05-14 | Verizon Patent And Licensing Inc. | Automatic media summary creation systems and methods |
KR20240050468A (ko) * | 2019-01-18 | 2024-04-18 | 스냅 아이엔씨 | 템플릿-기반 개인화 비디오 생성 시스템 및 방법 |
US11114086B2 (en) * | 2019-01-18 | 2021-09-07 | Snap Inc. | Text and audio-based real-time face reenactment |
-
2021
- 2021-04-08 KR KR1020210045651A patent/KR102345729B1/ko active IP Right Grant
- 2021-11-26 WO PCT/KR2021/017653 patent/WO2022215823A1/ko active Application Filing
- 2021-11-26 CN CN202180099111.0A patent/CN117441192A/zh active Pending
- 2021-11-26 JP JP2023562502A patent/JP2024513515A/ja not_active Withdrawn
- 2021-12-27 KR KR1020210188883A patent/KR20220139781A/ko not_active Application Discontinuation
-
2022
- 2022-04-05 US US17/713,465 patent/US20220328070A1/en not_active Abandoned
- 2022-04-05 EP EP22166669.6A patent/EP4071760A1/en not_active Withdrawn
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140218373A1 (en) * | 1999-05-21 | 2014-08-07 | Quonsil Pl. 3, Llc | Script control for camera positioning in a scene generated by a computer rendering engine |
KR20190046371A (ko) * | 2017-10-26 | 2019-05-07 | 에스케이텔레콤 주식회사 | 얼굴 표정 생성 장치 및 방법 |
KR20190087230A (ko) * | 2018-01-16 | 2019-07-24 | 삼성전자주식회사 | 복수의 카메라를 이용하여 영상 데이터를 생성하는 방법 및 서버 |
KR20200017266A (ko) * | 2018-08-08 | 2020-02-18 | 삼성전자주식회사 | 아바타에 대응하는 속성에 따른 아이템을 제공하는 방법 및 장치 |
KR20200046417A (ko) * | 2018-10-24 | 2020-05-07 | 삼성전자주식회사 | 보정 이미지 기반 아바타 생성 방법 및 장치 |
Also Published As
Publication number | Publication date |
---|---|
US20220328070A1 (en) | 2022-10-13 |
JP2024513515A (ja) | 2024-03-25 |
KR102345729B1 (ko) | 2022-01-03 |
KR20220139781A (ko) | 2022-10-17 |
CN117441192A (zh) | 2024-01-23 |
EP4071760A1 (en) | 2022-10-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022215823A1 (ko) | 영상 생성 방법 및 장치 | |
WO2019164266A1 (en) | Electronic device for generating image including 3d avatar reflecting face motion through 3d avatar corresponding to face and method of operating same | |
WO2017122900A1 (en) | Apparatus and method for operating personal agent | |
WO2020032597A1 (en) | Apparatus and method for providing item according to attribute of avatar | |
WO2020153785A1 (ko) | 전자 장치 및 이를 이용한 감정 정보에 대응하는 그래픽 오브젝트를 제공하는 방법 | |
WO2019013517A1 (en) | APPARATUS AND METHOD FOR VOICE CONTROL CONTEXT | |
US11657557B2 (en) | Method and system for generating data to provide an animated visual representation | |
CN110475069B (zh) | 图像的拍摄方法及装置 | |
WO2018054056A1 (zh) | 一种互动式运动方法及头戴式智能设备 | |
WO2020159288A1 (ko) | 전자 장치 및 그 제어 방법 | |
WO2020171541A1 (ko) | 전자 장치 및 이를 이용한 카메라 기능과 연동하여 이모지 편집을 위한 사용자 인터페이스를 제공하는 방법 | |
WO2021235636A1 (ko) | 사용자의 플레잉에 기초하여 재프로그래밍되는 인터랙티브 콘텐츠 제공 방법 및 장치 | |
KR102148151B1 (ko) | 디지털 커뮤니케이션 네트워크에 기반한 지능형 채팅 | |
EP3195601A1 (en) | Method of providing visual sound image and electronic device implementing the same | |
WO2018139203A1 (ja) | 情報処理装置、情報処理方法、及びプログラム | |
WO2021242005A1 (ko) | 전자 장치 및 사용자 아바타 기반의 이모지 스티커를 생성하는 방법 | |
CN111670431A (zh) | 信息处理装置、信息处理方法以及程序 | |
WO2022019692A1 (ko) | 애니메이션을 저작하기 위한 방법, 시스템 및 비일시성의 컴퓨터 판독 가능 기록 매체 | |
WO2024071884A1 (ko) | 민머리 인물 이미지 생성 장치 및 방법, 및 민머리 인물 이미지 생성 장치를 포함하는 가상 헤어 스타일링 체험 장치 및 이를 이용한 가상 헤어 스타일링 방법 | |
WO2021096339A1 (ko) | 이미지 변형 방법 | |
WO2018026082A1 (ko) | 애니메이션 제작 장치 및 방법 | |
WO2018174311A1 (ko) | 얼굴 인식 카메라의 동적 컨텐츠를 제공하는 방법 및 시스템 | |
CN117351123A (zh) | 交互式数字人像生成方法、装置、设备及存储介质 | |
WO2020122513A1 (ko) | 2차원 이미지 처리 방법 및 이 방법을 실행하는 디바이스 | |
JP2023099309A (ja) | アバターを通じて映像の音声を手話に通訳する方法、コンピュータ装置、およびコンピュータプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21936161 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023562502 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180099111.0 Country of ref document: CN |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21936161 Country of ref document: EP Kind code of ref document: A1 |