(19)
(11)EP 3 312 702 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
04.12.2019 Bulletin 2019/49

(21)Application number: 17196962.9

(22)Date of filing:  18.10.2017
(51)Int. Cl.: 
G06F 3/01  (2006.01)
G06K 9/00  (2006.01)

(54)

METHOD AND DEVICE FOR IDENTIFYING GESTURE

VERFAHREN UND VORRICHTUNG ZUR IDENTIFIZIERUNG EINER GESTE

PROCÉDÉ ET DISPOSITIF D'IDENTIFICATION DE GESTES


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 19.10.2016 CN 201610913061

(43)Date of publication of application:
25.04.2018 Bulletin 2018/17

(73)Proprietor: Beijing Xiaomi Mobile Software Co., Ltd.
Beijing 100085 (CN)

(72)Inventor:
  • LIN, Xingsheng
    Beijing, 100085 (CN)

(74)Representative: Grunert, Marcus 
Dehns Germany Theresienstraße 6-8
80333 München
80333 München (DE)


(56)References cited: : 
US-A1- 2013 182 902
US-A1- 2015 062 003
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    TECHNICAL FIELD



    [0001] The present invention generally relates to virtual reality technology field, and more particularly, to a method and a device for identifying a gesture.

    BACKGROUND



    [0002] With rapid development of virtual reality (VR) technology based on mobile apparatuses, a user can realize application functions under VR environment using a VR headset in combination with a mobile apparatus. For example, the user can play online games in the mobile apparatus using the VR headset. Human-computer interaction operation is usually involved in the process of using the VR headset.

    [0003] In the related art, information input by the user can be identified by gesture identification. In the gesture identification process, the mobile apparatus needs to analyze and process images collected by a camera, so as to determine a position of a hand of a target user, and then determine a gesture of the user according to the hand. Computation in image processing in this technical solution is very large, and when there are one or more hands in an image, the mobile apparatus may not be able to correctly identify the hand of the target user, such that input of effective information cannot be realized.
    US 2015/062003 A1 discloses methods and systems enabling users to interact using natural gestures with images created and displayed by electronic devices. An acquisition system preferably comprises at least two two-dimensional cameras. Image planes can be intersected to determine the three-dimensional position. A target region in each of the plurality of images is determined to be processed according to the depth. A gesture of a target user according to image information of target regions is determined.

    SUMMARY



    [0004] According to a first aspect of the present invention, there is provided a method for identifying a gesture according to claim 1

    [0005] In an embodiment, determining the depth of each pixel in each of the plurality images to be processed includes: determining camera parameters of the plurality of cameras; and calculating the depth of each pixel in each of the plurality of images to be processed according to the camera parameters of the plurality of cameras.

    [0006] According to a second aspect of the present invention, there is provided a device for identifying a gesture according to claim 3.

    [0007] In an embodiment, the depth determining module includes: a parameter determining sub module, configured to determine camera parameters of the plurality of cameras; and a depth determining sub module, configured to calculate the depth of each pixel in each of the plurality of images to be processed according to the camera parameters of the plurality of cameras which are determined by the parameter determining sub module.

    [0008] According to a third aspect of the present invention, there is provided a computer-readable storage medium having stored therein instructions that, when executed by a processor of a mobile apparatus comprising a plurality of cameras, causes the mobile apparatus to perform the above method according to the present invention, and a corresponding computer program.

    [0009] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present invention, as claimed.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0010] The accompanying drawings, which are incorporated in and constitute a part
    of this specification, illustrate embodiments consistent with the present invention and, together with the description, serve to explain the principles of the present invention.

    Fig. 1A is a flow chart illustrating a method for identifying a gesture, according to an exemplary embodiment.

    Fig. 1B is a schematic diagram illustrating determining a depth of each pixel with two cameras, according to an exemplary embodiment.

    Fig. 2 is a flow chart illustrating determining a target region according to the depth, according to a first exemplary embodiment.

    Fig. 3 is a flow chart illustrating determining a target depth by querying a characteristic database, according to a second exemplary embodiment.

    Fig. 4 is a flow chart illustrating determining a target depth according to history image data, according to a third exemplary embodiment.

    Fig. 5 is a block diagram illustrating a device for identifying a gesture, according to an exemplary embodiment.

    Fig. 6 is a block diagram illustrating a device for identifying a gesture, according to another exemplary embodiment.

    Fig. 7 is a block diagram illustrating a device for identifying a gesture, according to yet another exemplary embodiment.

    Fig. 8 is a block diagram illustrating an apparatus for identifying a gesture, according to an exemplary embodiment.


    DETAILED DESCRIPTION



    [0011] Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements unless otherwise represented. The implementations set forth in the following description of exemplary embodiments do not represent all implementations consistent with the present invention. Instead, they are merely examples of apparatuses and methods consistent with aspects related to the present invention as recited in the appended claims.

    [0012] Fig. 1A is a flow chart illustrating a method for identifying a gesture, according to an exemplary embodiment. Fig. 1B is a schematic diagram illustrating determining a depth of each pixel with two cameras, according to an exemplary embodiment. The method for identifying a gesture may be applied to a mobile apparatus with a plurality of cameras, for example, may be applied to a smartphone or a tablet computer. As illustrated in Fig. 1A, the method for identifying a gesture includes following acts.

    [0013] In block 101, a depth of each pixel in each of images to be processed is determined. The images to be processed are collected by the plurality of cameras separately. For example, the images to be processed are respectively collected by the plurality of cameras.

    [0014] In one or more embodiments, the depth is configured to at least partially represent a distance between an actual object point corresponding to each pixel and the mobile apparatus.

    [0015] For example, as illustrated in Fig. 1B, a first camera O1 of a mobile apparatus 110 photographs an actual object point A on an image plane 120, and a second camera O2 of the mobile apparatus 110 photographs the actual object point A on an image plane 130. A distance BC between the actual object point A and the mobile apparatus 110 can be calculated according to a focal length of the first camera O1, a focal length of the second camera O2, and a trigonometric relationship in Fig. 1B.

    [0016] In block 102, an image region to be processed of each of the images to be processed is determined according to the depth. The image region to be processed may also be referred as the target region.

    [0017] In one or more embodiments, a method for determining the target region of each of the images to be processed according to the depth may refer to embodiments illustrated in Fig. 2, which is not described in detail herein.

    [0018] For example, the target region is configured to represent a region for identifying a gesture. The camera may collect a lot of invalid background information in actual collection of hand actions of a target user. For example, a background behind hands of the target user, such as a wall. The invalid background information can be marked as an image region not to be processed according to the depth of each pixel in the images to be processed, thus reducing computation in image processing.

    [0019] In block 103, a gesture of the target user is determined according to image information of image regions to be processed.

    [0020] In this embodiment, the mobile apparatus with the plurality of cameras can determine the depth of each pixel in each of the collected images to be processed, and determine the target region which contains hand image of the target user according to the depth of each pixel, and then process the image regions to be processed so as to realize gesture identification of the target user, such that a problem of large computation in image processing caused by processing the whole image to be processed in the related art is solved, and a problem that input of effective information cannot be realized when there are one or more hands in the image in the related art is solved, thus improving user experience in virtual reality when the user inputs information.

    [0021] In one or more embodiments, determining a depth of each pixel in each of the images to be processed may include following acts. Camera parameters of the plurality of cameras are determined. The depth of each pixel in each of the images to be processed is calculated according to the camera parameters of the plurality of cameras.

    [0022] In one or more embodiments, determining a target region of each of the images to be processed according to the depth may include following acts. The depth of each pixel is compared with a target depth. The target depth is configured to represent a maximum distance between a hand of the target user and the mobile apparatus. A pixel having depth less than the target depth is determined as a pixel to be processed in each of the images to be processed. A region composed of pixels to be processed is determined as the target region.

    [0023] In one or more embodiments, the method may further include following acts. A first physical characteristic of the target user is determined by querying a first characteristic database. The first characteristic database is configured to record user identification information and first physical characteristics corresponding to the user identification information. The target depth is calculated according to the first physical characteristic. The second physical characteristic may also be referred as the additional physical characteristic.

    [0024] In an embodiment, the method may further include following acts. A second physical characteristic of the target user is determined by querying a second characteristic database. The second characteristic database is configured to record user identification information and second physical characteristics corresponding to the user identification information. A first physical characteristic of the target user is determined according the second physical characteristic. The target depth is calculated according to the first physical characteristic.

    [0025] In an embodiment, the method may further include following acts. A plurality of reference distances between the hand of the target user and the mobile apparatus is determined according to history image data collected by the plurality of cameras separately. A maximum of the plurality of reference distances is determined. The maximum of the plurality of reference distances is determined as the target depth..

    [0026] Details on how to identify a gesture can refer to following embodiments.

    [0027] Technical solutions provided by the present invention will be described below with reference to following embodiments.

    [0028] Fig. 2 is a flow chart illustrating determining a target region according to the depth, according to a first exemplary embodiment. In this embodiment, by using the method provided in the above embodiments of the present invention, and by taking a mobile apparatus with a plurality of cameras as an example, an exemplary illustration of determining a target region is made. As illustrated in Fig. 2, determining a target region according to the depth includes following acts.

    [0029] In block 201, the depth of each pixel is compared with a target depth.

    [0030] In one or more embodiments, the target depth is configured to represent a maximum distance between a hand of the target user and the mobile apparatus.

    [0031] In an embodiment, the target depth can be obtained according to examples illustrated in Fig. 3 and Fig. 4.

    [0032] In block 202, a pixel having depth that is less than the target depth is determined as a pixel to be processed in each of the images to be processed.

    [0033] In block 203, a region composed of pixels to be processed is determined as the target region.

    [0034] In one or more embodiments, by comparing the depth of each pixel with the target depth, effective pixels to be processed can be determined, and then the target region is determined. Because the target depth is configured to represent the maximum distance between the hand of the target user and the mobile apparatus, the target region which is determined according to the target depth represents a region containing the hand of the target user, and a region other than the target region is a region where it is impossible to contain the hand of the target user. By analyzing and identifying the target region only, computation in image processing is reduced, and the problem that a hand action occurring in the region outside the target region is identified as the gesture of the target user is avoided, thus realizing input of effective information.

    [0035] Fig. 3 is a flow chart illustrating determining a target depth by querying a characteristic database, according to a second exemplary embodiment. In this embodiment, by using the method provided in the above embodiments of the present invention, and by taking a mobile apparatus with a plurality of cameras as an example, an exemplary illustration of determining a target depth by querying a characteristic database is made. As illustrated in Fig. 3, determining a target depth by querying a characteristic database includes following acts.

    [0036] In block 301, a first physical characteristic of the target user is determined.

    [0037] In one or more embodiments, the first physical characteristic may be a length of arms of a user. For example, the first physical characteristic is 75 centimeter.

    [0038] In an embodiment, the first physical characteristic of the target user can be determined by querying a first characteristic database according to a user account of the target user. For example, when the user is operating an application in the mobile apparatus, such as playing an online game named "Affected The Manor", the user may be requested to provide his first physical characteristic. Then his first physical characteristic is stored in the first characteristic database. When the user plays online games with a VR helmet later, the first physical characteristic of the user can be determined according to a login account of the user.

    [0039] In an embodiment, a second physical characteristic of the target user can be determined by querying a second characteristic database according to a user account of the target user. For example, the second physical characteristic may be a height of a user. Then the first physical characteristic of the target user can be determined according to the second physical characteristic. For example, a length of arms of a user with a height of 1.7 meter is usually 75-80 centimeter, such that the length of arms of the user can be marked as a maximum of the range from 75 to 80 centimeter, i.e., 80 centimeter.

    [0040] For example, after the second physical characteristic of the target user is determined, a first physical characteristic corresponding to the second physical characteristic can be determined by accessing a network server. In another embodiment, a characteristic matching list may be realized locally, and the characteristic matching list records first physical characteristics corresponding to second physical characteristics respectively.

    [0041] In block 302, the target depth is calculated according to the first physical characteristic.

    [0042] In one or more embodiments, the target depth can be calculated according to formula (1),

    where, L represents the target depth, w represents a weight coefficient, and T represents a characteristic value of the first physical characteristic.

    [0043] For example, the weight coefficient w can be obtained by analyzing a using habit of a user in a period when the user initially uses a VR device (for example, a month when the VR device is initially used). For example, some users are accustomed to straighten their arms when they making gestures, then the weight coefficient w may be larger for these users, 0.9 for example, while some users are accustomed to bend their arms when they making gestures, then the weight coefficient w may be smaller for these users, 0.6 for example.

    [0044] In this embodiment, when determining the target depth, the first physical characteristic (such as the length of arm) of the target user can be obtained by querying the first characteristic database, and then the target depth is calculated; alternatively, the second physical characteristic (such as the height) of the target user can be obtained by querying the second characteristic database, then a first physical characteristic is determined according to the second physical characteristic, and then the target depth is calculated, such that the target depth can be determined flexibly, thus improving the user experience in virtual reality when the user inputs information.

    [0045] Fig. 4 is a flow chart illustrating determining a target depth according to history image data, according to a third exemplary embodiment. In this embodiment, by using the method provided in the above embodiments of the present invention, and by taking a mobile apparatus with a plurality of cameras as an example, an exemplary illustration of determining a target depth according to history image data is made. As illustrated in Fig. 4, determining a target depth according to history image data includes following acts.

    [0046] In block 401, a plurality of reference distances between the hand of the target user and the mobile apparatus is determined according to history image data collected by the plurality of cameras separately.

    [0047] In one or more embodiments, the plurality of reference distances between the hand of the target user and the mobile apparatus can be determined by analyzing history image data in a period when the user initially uses a VR device (for example, a week when the VR device is initially used). For example, a depth when the user performs some gestures is 0.6 meter, and a depth when the user performs some other gestures is 0.55 meter, then the plurality of reference distances can be obtained by analyzing different image data.

    [0048] In block 402, a maximum of the plurality of reference distances is determined.

    [0049] In block 403, the maximum of the plurality of reference distances is determined as the target depth, such that the act that the depth of each pixel is compared with a target depth can be executed based on the target depth.

    [0050] In one or more embodiments, the reference distances between the hand of the target user and the mobile apparatus can be analyzed according to the history image data, and then the target depth is calculated, thus making the target depth more in line with the user's actual physical characteristics and using habits. For example, for two users with an arm length of 60 centimeter, a difference between target depths of the two users may be 10 -20 centimeter because of different habits for stretching arms. Therefore, the target depth determined by analyzing history image data can adapt to the target user better.

    [0051] Fig. 5 is a block diagram illustrating a device for identifying a gesture, according to an exemplary embodiment. The device for identifying a gesture is applied to a mobile apparatus with a plurality of cameras. As illustrated as Fig. 5, the device for identifying a gesture includes a depth determining module 510, a region determining module 520, and a gesture identifying module 530.

    [0052] The depth determining module 510 is configured to determine a depth of each pixel in each of images to be processed. The images to be processed are collected by the plurality of cameras separately. The depth is configured to at least partially represent a distance between an actual object point corresponding to each pixel and the mobile apparatus.

    [0053] The region determining module 520 is configured to determine a target region of each of the images to be processed according to the depth determined by the depth determining module 510.

    [0054] The gesture identifying module 530 is configured to determine a gesture of a target user according to image information of image regions to be processed which are determined by the region determining module 520.

    [0055] Fig. 6 is a block diagram illustrating a device for identifying a gesture, according to another exemplary embodiment. As illustrated as Fig. 6, based on embodiments shown in Fig. 5, in an embodiment, the depth determining module 510 includes a parameter determining sub module 511 and a depth determining sub module 512.

    [0056] The parameter determining sub module 511 is configured to determine camera parameters of the plurality of cameras.

    [0057] The depth determining sub module 512 is configured to calculate the depth of each pixel in each of the images to be processed according to the camera parameters of the plurality of cameras which are determined by the parameter determining sub module 511.

    [0058] In some embodiments, the region determining module 520 includes a comparing sub module 521, a pixel determining sub module 522, and a region determining sub module 523.

    [0059] The comparing sub module 521 is configured to compare the depth of each pixel with a target depth. The target depth is configured to represent a maximum distance between a hand of the target user and the mobile apparatus.

    [0060] The pixel determining sub module 522 is configured to determine a pixel having a depth less than the target depth as a pixel to be processed in each of the images to be processed.

    [0061] The region determining sub module 523 is configured to determine a region composed of pixels to be processed which are determined by the pixel determining sub module 522 as the target region.

    [0062] Fig. 7 is a block diagram illustrating a device for identifying a gesture, according to yet another exemplary embodiment. As shown in Fig. 7, based on examples shown in Fig. 5 and Fig. 6, the device may further include a first querying module 540 and a first calculating module 550.

    [0063] The first querying module 540 is configured to determine a first physical characteristic of the target user by querying a first characteristic database. The first characteristic database is configured to record user identification information and first physical characteristics corresponding to the user identification information.

    [0064] The first calculating module 550 is configured to calculate the target depth according to the first physical characteristic determined by the first querying module 540, such that the comparing sub module 521 compares the depth of each pixel with the target depth.

    [0065] In some embodiments, the device may further include a second querying module 560, a characteristic determining module 570 and a second calculating module 580.

    [0066] The second querying module 560 is configured to determine a second physical characteristic of the target user by querying a second characteristic database. The second characteristic database is configured to record user identification information and second physical characteristics corresponding to the user identification information

    [0067] The characteristic determining module 570 is configured to determine a first physical characteristic of the target user according the second physical characteristic determined by the second querying module 560.

    [0068] The second calculating module 580 is configured to calculate the target depth according to the first physical characteristic determined by the characteristic determining module 570, such that the comparing sub module 621 compares the depth of each pixel with the target depth.

    [0069] In an embodiment, the device may further include a referring module 590, a maximum module 600, and a target depth determining module 610.

    [0070] The referring module 590 is configured to determine a plurality of reference distances between the hand of the target user and the mobile apparatus according to history image data collected by the plurality of cameras separately.

    [0071] The maximum module 600 is configured to determine a maximum of the plurality of reference distances determined by the referring module 590.

    [0072] The target depth determining module 610 is configured to determine the maximum of the plurality of reference distances as the target depth, such that the comparing sub module 521 compares the depth of each pixel with the target depth.

    [0073] With respect to the devices in the above embodiments, the specific manners for performing operations for individual modules therein have been described in detail in the embodiments regarding the methods for identifying a gesture, which will not be elaborated herein.

    [0074] Since the device embodiments substantially correspond to the method embodiments, reference is made to the description of the method embodiments. The above-described device embodiments are merely for the purpose of illustration. Those units described as separated components may or may not be physically separated, and those units described as a display component may or may not be physical units, i.e., either located at one place, or distributed onto a plurality of network units. The object of the present invention may be achieved by part or all of modules in accordance with practical requirements. It would be appreciated and executable by those skilled in the art without creative labor.

    [0075] Fig. 8 is a block diagram illustrating an apparatus for identifying a gesture, according to yet another exemplary embodiment. For example, the device 800 may be a mobile apparatus, such as a smart phone, a tablet.

    [0076] Referring to Fig. 8, the device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.

    [0077] The processing component 802 typically controls overall operations of the device 800, such as the operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to perform all or part of the steps in the above described methods. Moreover, the processing component 802 may include one or more modules which facilitate the interaction between the processing component 802 and other components. For instance, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

    [0078] The memory 804 is configured to store various types of data to support the operation of the device 800. Examples of such data include instructions for any applications or methods operated on the device 800, contact data, phonebook data, messages, pictures, video, etc. The memory 804 may be implemented using any type of volatile or non-volatile memory devices, or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic or optical disk.

    [0079] The power component 806 provides power to various components of the device 800. The power component 806 may include a power management system, one or more power sources, and any other components associated with the generation, management, and distribution of power in the device 800.

    [0080] The multimedia component 808 includes a screen providing an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes the touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may not only sense a boundary of a touch or swipe action, but also sense a period of time and a pressure associated with the touch or swipe action.

    [0081] The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a microphone ("MIC") configured to receive an external audio signal when the device 800 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may be further stored in the memory 804 or transmitted via the communication component 816. In some embodiments, the audio component 810 further includes a speaker to output audio signals.

    [0082] The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, such as a keyboard, a click wheel, buttons, and the like. The buttons may include, but are not limited to, a home button, a volume button, a starting button, and a locking button.

    [0083] The sensor component 814 includes one or more sensors to provide status assessments of various aspects of the device 800. For instance, the sensor component 814 may detect an open/closed status of the device 800, relative positioning of components, e.g., the display and the keypad, of the device 800, a change in position of the device 800 or a component of the device 800, a presence or absence of user contact with the device 800, an orientation or an acceleration/deceleration of the device 800, and a change in temperature of the device 800. The sensor component 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor component 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 814 may also include an accelerometer sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

    [0084] The communication component 816 is configured to facilitate communication, wired or wirelessly, between the device 800 and other devices. The device 800 can access a wireless network based on a communication standard, such as WiFi, 2G, or 3G, or a combination thereof. In one exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast associated information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component 816 further includes a near field communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on a radio frequency identification (RFID) technology, an infrared data association (IrDA) technology, an ultra-wideband (UWB) technology, a Bluetooth (BT) technology, and other technologies.

    [0085] In exemplary embodiments, the device 800 may be implemented with one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic components, for performing the above described methods.

    [0086] In exemplary embodiments, there is also provided a non-transitory computer-readable storage medium including instructions, such as included in the memory 804, executable by the processor 820 in the device 800, for performing the above-described methods. For example, the non-transitory computer-readable storage medium may be a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disc, an optical data storage device, and the like.

    [0087] Other embodiments of the present invention will be apparent to those skilled in the art from consideration of the specification and practice of the present invention disclosed here.


    Claims

    1. A method for identifying a gesture, applied in a mobile apparatus comprising a plurality of cameras (O1, O2), the method comprising:

    determining (101) a depth of each pixel in a plurality of images to be processed, wherein the plurality of images to be processed are collected by the plurality of cameras, and the depth is configured to at least partially represent a distance between an actual object point corresponding to each pixel and the mobile apparatus;

    determining (102) a target region in each of the plurality of images to be processed according to the depth; and

    determining (103) a gesture of a target user according to image information of target regions,

    wherein determining (102) the target region in each of the plurality of images to be processed according to the depth comprises:

    comparing (201) the depth of each pixel with a target depth, wherein the target depth is configured to represent a maximum distance between a hand of the target user and the mobile apparatus;

    determining (202) a pixel having a depth less than the target depth as a pixel to be processed in each of the plurality of images to be processed; and

    determining (203) a region composed of pixels to be processed as the target region;

    characterized in that the method further comprises:

    determining (401) a plurality of reference distances between the hand of the target user and the mobile apparatus according to history image data collected separately by the plurality of cameras;

    determining (402) a maximum of the plurality of reference distances; and

    determining (403) the maximum of the plurality of reference distances as the target depth.


     
    2. The method according to claim 1, wherein determining (101) the depth of each pixel in the plurality of images to be processed comprises:

    determining camera parameters of the plurality of cameras; and

    calculating the depth of each pixel in each of the plurality of images to be processed according to the camera parameters of the plurality of cameras.


     
    3. A device for identifying a gesture in a mobile (110) apparatus comprising a plurality of cameras (O1, O2), the device comprising:

    a depth determining module (510), configured to determine a depth of each pixel in a plurality of images (120, 130) to be processed, wherein the plurality of images to be processed are collected by the plurality of cameras, and the depth is configured to at least partially represent a distance between an actual object point corresponding to each pixel and the mobile apparatus;

    a region determining module (520), configured to determine a target region of each of the plurality of images to be processed according to the depth determined by the depth determining module (510); and

    a gesture identifying module (530), configured to determine a gesture of a target user according to image information of target regions determined by the region determining module (520),

    wherein the region determining module (520) comprises:

    a comparing sub module (521), configured to compare the depth of each pixel which is determined by the depth determining module with a target depth, wherein the target depth is configured to represent a maximum distance between a hand of the target user and the mobile apparatus;

    a pixel determining sub module (522), configured to determine a pixel having a depth less than the target depth which is determined by the comparing sub module (521) as a pixel to be processed in each of the plurality of images to be processed; and

    a region determining sub module (523), configured to determine a region composed of pixels to be processed which are determined by the pixel determining sub module (522) as the target region;

    characterized in that the device further comprises:

    a referring module (590), configured to determine a plurality of reference distances between the hand of the target user and the mobile apparatus according to history image data collected separately by the plurality of cameras;

    a maximum module (600), configured to determine a maximum of the plurality of reference distances; and

    a target depth determining module (610), configured to determine the maximum of the plurality of reference distances as the target depth, such that the comparing sub module (521) compares the depth of each pixel with the target depth.


     
    4. The device according to claim 3, wherein the depth determining module (510) comprises:

    a parameter determining sub module (511), configured to determine camera parameters of the plurality of cameras; and

    a depth determining sub module (512), configured to calculate the depth of each pixel in each of the plurality of images to be processed according to the camera parameters of the plurality of cameras which are determined by the parameter determining sub module (511).


     
    5. A computer-readable storage medium having stored therein instructions that, when executed by a processor of a mobile apparatus comprising a plurality of cameras (O1, O2), causes the mobile apparatus to perform the method according to any one of claims 1 or 2.
     
    6. A computer program including instructions that, when executed by a processor of a mobile apparatus comprising a plurality of cameras (O1, O2), causes the mobile apparatus to perform the method according to any one of claims 1 or 2.
     


    Ansprüche

    1. Verfahren zum Identifizieren einer Geste, die in einer mobilen Vorrichtung angewendet wird, die eine Vielzahl von Kameras (O1, O2) umfasst, wobei das Verfahren umfasst:

    Bestimmen (101) einer Tiefe jedes Pixels in einer Vielzahl von zu verarbeitenden Bildern, wobei die Vielzahl von zu verarbeitenden Bildern von der Vielzahl von Kameras gesammelt wird, und wobei die Tiefe eingerichtet ist, zumindest teilweise einen Abstand zwischen einem aktuellen bzw. tatsächlichen Objektpunkt, der jedem Pixel entspricht, und der mobilen Vorrichtung darzustellen;

    Bestimmen (102) eines Zielbereichs in jedem der Vielzahl von zu verarbeitenden Bildern entsprechend der Tiefe; und

    Bestimmen (103) einer Geste eines Zielanwenders gemäß Bildinformationen von Zielbereichen,

    wobei das Bestimmen (102) des Zielbereichs in jedem der Vielzahl von zu verarbeitenden Bildern entsprechend der Tiefe umfasst:

    Vergleichen (201) der Tiefe jedes Pixels mit einer Zieltiefe, wobei die Zieltiefe eingerichtet ist, einen maximalen Abstand zwischen einer Hand des Zielanwenders und der mobilen Vorrichtung darzustellen;

    Bestimmen (202) eines Pixels, der eine Tiefe aufweist, die kleiner ist als die Zieltiefe, als ein Pixel, der in jedem der Vielzahl von zu verarbeitenden Bildern verarbeitet werden soll; und

    Bestimmen (203) eines Bereichs, der aus Pixeln gebildet ist, die als Zielbereich verarbeitet werden sollen;

    dadurch gekennzeichnet, dass das Verfahren ferner umfasst:

    Bestimmen (401) einer Vielzahl von Referenzabständen zwischen der Hand des Zielanwenders und der mobilen Vorrichtung entsprechend den Verlaufsbilddaten, die separat von der Vielzahl von Kameras gesammelt werden;

    Bestimmen (402) eines Maximums der Vielzahl von Referenzabständen; und

    Bestimmen (403) des Maximums der Vielzahl von Referenzabständen als die Zieltiefe.


     
    2. Verfahren nach Anspruch 1, wobei das Bestimmen (101) der Tiefe jedes Pixels in der Vielzahl der zu verarbeitenden Bilder umfasst:

    Bestimmen von Kameraparametern der Vielzahl von Kameras; und

    Berechnen der Tiefe jedes Pixels in jedem der Vielzahl von zu verarbeitenden Bildern entsprechend den Kameraparametern der Vielzahl von Kameras.


     
    3. Vorrichtung zum Identifizieren einer Geste in einer mobilen (110) Vorrichtung, die eine Vielzahl von Kameras (O1, O2) umfasst, wobei die Vorrichtung umfasst:

    ein Tiefenbestimmungsmodul (510), das eingerichtet ist, eine Tiefe jedes Pixels in einer Vielzahl von zu verarbeitenden Bildern (120, 130) zu bestimmen, wobei die Vielzahl von zu verarbeitenden Bildern von der Vielzahl von Kameras gesammelt wird, und wobei die Tiefe eingerichtet ist, zumindest teilweise einen Abstand zwischen einem tatsächlichen Objektpunkt, der jedem Pixel entspricht, und der mobilen Vorrichtung darzustellen;

    ein Bereichsbestimmungsmodul (520), das eingerichtet ist, einen Zielbereich von jedem der Vielzahl von zu verarbeitenden Bildern gemäß der durch das Tiefenbestimmungsmodul (510) bestimmten Tiefe zu bestimmen; und

    ein Gestenidentifikationsmodul (530), das eingerichtet ist, eine Geste eines Zielanwenders entsprechend den Bildinformationen von Zielbereichen zu bestimmen, die durch das das Bereichsbestimmungsmodul (520) bestimmt werden,

    wobei das Bereichsbestimmungsmodul (520) umfasst:

    ein Vergleichssubmodul (521), das eingerichtet ist, die Tiefe jedes Pixels, die durch das Tiefenbestimmungsmodul bestimmt wird, mit einer Zieltiefe zu vergleichen, wobei die Zieltiefe eingerichtet ist, einen maximalen Abstand zwischen einer Hand des Zielanwenders und der mobilen Vorrichtung darzustellen;

    ein Pixelbestimmungssubmodul (522), das eingerichtet ist, ein Pixel mit einer Tiefe zu bestimmen, die kleiner ist als die Zieltiefe, die durch das Vergleichssubmodul (521) als ein Pixel bestimmt wird, der in jedem der Vielzahl der zu verarbeitenden Bilder verarbeitet werden soll; und

    ein Bereichsbestimmungssubmodul (523), das eingerichtet ist, einen Bereich zu bestimmen, der aus zu verarbeitenden Pixeln gebildet ist, die durch das Pixelbestimmungssubmodul (522) als der Zielbereich bestimmt werden;

    dadurch gekennzeichnet, dass die Vorrichtung ferner umfasst:

    ein Referenzmodul (590), das eingerichtet ist, eine Vielzahl von Referenzabständen zwischen der Hand des Zielanwenders und der mobilen Vorrichtung entsprechend den Verlaufsbilddaten zu bestimmen, die separat von der Vielzahl von Kameras gesammelt werden;

    ein Maximummodul (600), das eingerichtet ist, ein Maximum der Vielzahl von Referenzabständen zu bestimmen; und

    ein Zieltiefenbestimmungsmodul (610), das eingerichtet ist, das Maximum der Vielzahl von Referenzabständen als die Zieltiefe zu bestimmen, so dass das Vergleichssubmodul (521) die Tiefe jedes Pixels mit der Zieltiefe vergleicht.


     
    4. Vorrichtung nach Anspruch 3, wobei das Tiefenbestimmungsmodul (510) umfasst:

    ein Parameterbestimmungssubmodul (511), das eingerichtet ist, Kameraparameter der Vielzahl von Kameras zu bestimmen; und

    ein Tiefenbestimmungssubmodul (512), das eingerichtet ist, die Tiefe jedes Pixels in jedem der Vielzahl von zu verarbeitenden Bildern entsprechend den Kameraparametern der Vielzahl von Kameras zu berechnen, die durch das Parameterbestimmungssubmodul (511) bestimmt werden.


     
    5. Computerlesbares Speichermedium, auf bzw. in dem Anweisungen gespeichert sind, die, wenn sie von einem Prozessor einer mobilen Vorrichtung mit einer Vielzahl von Kameras (O1, O2) ausgeführt werden, bewirken, dass die mobile Vorrichtung das Verfahren nach einem der Ansprüche 1 oder 2 durchführt.
     
    6. Computerprogramm mit Anweisungen, die, wenn sie von einem Prozessor einer mobilen Vorrichtung mit einer Vielzahl von Kameras (O1, O2) ausgeführt werden, bewirken, dass die mobile Vorrichtung das Verfahren nach einem der Ansprüche 1 oder 2 durchführt.
     


    Revendications

    1. Procédé d'identification d'un geste, appliqué dans un dispositif mobile comprenant une pluralité de caméras (O1, O2), le procédé comprenant :

    déterminer (101) une profondeur de chaque pixel dans une pluralité d'images à traiter, dans lequel la pluralité d'images à traiter sont collectées par la pluralité de caméras, et la profondeur est configurée pour représenter au moins partiellement une distance entre un point objet réel correspondant à chaque pixel et le dispositif mobile ;

    déterminer (102) une région cible dans chacune de la pluralité d'images à traiter en fonction de la profondeur ; et

    déterminer (103) un geste d'un utilisateur cible en fonction des informations d'image des régions cibles,

    dans laquelle la détermination (102) de la région cible dans chacune de la pluralité d'images à traiter en fonction de la profondeur comprend :

    comparer (201) la profondeur de chaque pixel avec une profondeur cible, dans lequel la profondeur cible est configurée pour représenter une distance maximale entre une main de l'utilisateur cible et le dispositif mobile ;

    déterminer (202) un pixel ayant une profondeur inférieure à la profondeur cible comme pixel à traiter dans chacune de la pluralité d'images à traiter ; et

    déterminer (203) une région composée de pixels à traiter comme région cible ;

    caractérisé en ce que le procédé comprend en outre :

    l'utilisateur cible et le dispositif mobile en fonction de données d'image historiques collectées séparément par la pluralité de caméras ;

    déterminer (402) un maximum de la pluralité de distances de référence ; et

    déterminer (403) le maximum de la pluralité de distances de référence comme profondeur cible.


     
    2. Procédé selon la revendication 1, dans lequel la détermination (101) de la profondeur de chaque pixel dans la pluralité d'images à traiter comprend :

    déterminer les paramètres de caméra de la pluralité de caméras ; et

    calculer la profondeur de chaque pixel dans chacune de la pluralité d'images à traiter en fonction des paramètres de caméra de la pluralité de caméras.


     
    3. Dispositif pour identifier un geste dans un dispositif mobile (110) comprenant une pluralité de caméras (O1, O2), le dispositif comprenant :

    un module de détermination de profondeur (510), configuré pour déterminer une profondeur de chaque pixel dans une pluralité d'images (120, 130) à traiter, dans lequel la pluralité d'images à traiter sont collectées par la pluralité de caméras, et la profondeur est configurée pour représenter au moins partiellement une distance entre un point objet réel correspondant à chaque pixel et le dispositif mobile ;

    un module de détermination de région (520), configuré pour déterminer une région cible de chacune de la pluralité d'images à traiter selon la profondeur déterminée par le module de détermination de profondeur (510) ; et

    un module d'identification de geste (530), configuré pour déterminer un geste d'un utilisateur cible selon des informations d'image de régions cibles déterminées par le module de détermination de région (520),

    dans laquelle le module de détermination de région (520) comprend :

    un sous-module de comparaison (521), configuré pour comparer la profondeur de chaque pixel qui est déterminée par le module de détermination de profondeur avec une profondeur cible, dans lequel la profondeur cible est configurée pour représenter une distance maximale entre une main de l'utilisateur cible et le dispositif mobile ;

    un sous-module de détermination de pixel (522), configuré pour déterminer un pixel ayant une profondeur inférieure à la profondeur cible qui est déterminée par le sous-module de comparaison (521) comme pixel à traiter dans chacune de la pluralité d'images à traiter ; et

    un sous-module de détermination de région (523), configuré pour déterminer une région composée de pixels à traiter qui sont déterminés par le sous-module de détermination de pixel (522) comme région cible ;

    caractérisé en ce que le dispositif comprend en outre :

    un module de référence (590), configuré pour déterminer une pluralité de distances de référence entre la main de l'utilisateur cible et le dispositif mobile en fonction de données d'image historiques collectées séparément par la pluralité de caméras ;

    un module maximum (600), configuré pour déterminer un maximum de la pluralité de distances de référence ; et

    un module de détermination de profondeur de cible (610), configuré pour déterminer le maximum de la pluralité de distances de référence comme profondeur de cible, de sorte que le sous-module de comparaison (521) compare la profondeur de chaque pixel avec la profondeur cible.


     
    4. Dispositif selon la revendication 3, dans lequel le module de détermination de profondeur (510) comprend :

    un sous-module de détermination de paramètres (511), configuré pour déterminer les paramètres de caméra de la pluralité de caméras ; et

    un sous-module de détermination de profondeur (512), configuré pour calculer la profondeur de chaque pixel dans chacune de la pluralité d'images à traiter selon les paramètres de caméra de la pluralité de caméras qui sont déterminés par le sous-module de détermination de paramètre (511).


     
    5. Support de stockage lisible par ordinateur dans lequel sont stockées des instructions qui, lorsqu'elles sont exécutées par un processeur d'un dispositif mobile comprenant une pluralité de caméras (O1, O2), amène le dispositif mobile à exécuter le procédé selon l'une quelconque des revendications 1 ou 2.
     
    6. Programme d'ordinateur comprenant des instructions qui, lorsqu'elles sont exécutées par un processeur d'un dispositif mobile comprenant une pluralité de caméras (O1, O2), amène le dispositif mobile à exécuter le procédé selon l'une quelconque des revendications 1 ou 2.
     




    Drawing















    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description