WO2020038140A1 - 人脸识别方法及装置 - Google Patents
人脸识别方法及装置 Download PDFInfo
- Publication number
- WO2020038140A1 WO2020038140A1 PCT/CN2019/095338 CN2019095338W WO2020038140A1 WO 2020038140 A1 WO2020038140 A1 WO 2020038140A1 CN 2019095338 W CN2019095338 W CN 2019095338W WO 2020038140 A1 WO2020038140 A1 WO 2020038140A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- face
- image
- rgb image
- target
- rgb
- Prior art date
Links
Images
Classifications
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- 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 present application relates to the field of computer technology, and in particular, to a face recognition method and device.
- swipe face can be applied in more and more scenarios, such as swipe face payment, swipe face to sign in, swipe face to unlock access control, swipe face authentication service, etc. Fast and so on.
- swipe face can be applied in more and more scenarios, such as swipe face payment, swipe face to sign in, swipe face to unlock access control, swipe face authentication service, etc. Fast and so on.
- a face recognition method comprising:
- RGB image and a corresponding depth image for face recognition, the RGB image including at least one face;
- a face recognition device includes:
- An acquisition module configured to acquire an RGB image and a corresponding depth image for face recognition, where the RGB image includes at least one face;
- a selection module configured to select a target face from the RGB image
- a judging module configured to judge whether there is an interfering face in the RGB image according to the target face and the depth image, and the distance from the interfering face to the face image acquisition device and the target face to The difference between the distances of the face image acquisition devices is less than a preset threshold;
- a recognition module configured to perform face recognition based on the target face when the interference face does not exist in the RGB image.
- an electronic device including:
- a memory arranged to store computer-executable instructions that, when executed, cause the processor to perform the following operations:
- RGB image and a corresponding depth image for face recognition, the RGB image including at least one face;
- a computer storage medium stores one or more programs, and the one or more programs, when executed by an electronic device including a plurality of application programs, cause the electronic The device does the following:
- RGB image and a corresponding depth image for face recognition, the RGB image including at least one face;
- a depth image corresponding to the RGB image may be combined to determine the usefulness of the RGB image.
- Face for face recognition since the depth image contains rich information, and the depth image can reflect the distance from each face in the depth image to the image acquisition device, and the face The distance to the image acquisition device can reflect the user's face recognition willingness to a certain extent. Therefore, the embodiments of this specification can avoid missing detection of the face in the RGB image, and can more accurately determine the use of the face in the RGB image. Recognized faces.
- FIG. 1 is a flowchart of a face recognition method according to an embodiment of the present specification
- FIG. 2 is a flowchart of a face recognition method according to another embodiment of the present specification.
- FIG. 3 is a schematic structural diagram of a face recognition device according to an embodiment of the present specification.
- FIG. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present specification.
- the embodiments of the present specification provide a face recognition method and device.
- the face recognition method provided in the embodiments of this specification is applicable to electronic devices.
- the electronic device may be a server, or the electronic device may also be a terminal such as a mobile phone, a tablet computer, or a personal digital assistant.
- Device or the electronic device may also be a computer device such as a notebook computer, a desktop computer, or a desktop computer, which is not limited in the embodiments of the present specification.
- FIG. 1 is a flowchart of a face recognition method according to an embodiment of the present specification. As shown in FIG. 1, the method may include the following steps: step 102, step 104, step 106, and step 108, where
- step 102 an RGB image and a corresponding depth image for face recognition are acquired, where the RGB image includes at least one face.
- the RGB image (color map) and corresponding depth image used for face recognition are images taken for the same scene.
- the gray value of each pixel in the depth image can be used to characterize the distance from a point in the shooting scene to the depth image acquisition device.
- the device used to capture depth images is called a depth image capture device, and the device used to capture RGB color images is called an RGB image capture device.
- step 104 a target face is selected from the RGB image.
- the target face image is the face most likely to be used for face recognition in the RGB image.
- face detection may be performed on an RGB image, a face contained therein is detected, and a face is selected from the face as a target face.
- a human face in a preset area in the RGB image may be selected as a target human face.
- the preset area may include: the central area of the RGB image, or Focus area when shooting RGB images.
- the face in the center area of the RGB image can be selected as the target face; or the face in the focus area when the RGB image is taken can be selected as the target face.
- step 106 according to the target face and the depth image, it is determined whether there is an interference face in the RGB image; if not, step 108 is performed; wherein the distance from the interference face to the face image acquisition device and the target face to the person The difference between the distances of the face image acquisition devices is less than a preset threshold.
- the face image acquisition device refers to a depth image acquisition device.
- the distance between the interference face and the target face to the depth image acquisition device is equal or similar.
- RGB image Interfering faces determine if the target face is the face with the most intent of face recognition in a multi-person scene; specifically, if there is an interfering face in an RGB image, it indicates that the target face is not the most popular in a multi-person scene The face with the intention of face recognition; if there is no disturbing face in the RGB image, it indicates that the target face has the most face recognition intention in the multi-person scene.
- RGB images may sometimes result in missed detection of faces, for example, faces in the corners of RGB images or half faces appearing in RGB images cannot be detected. Based on this situation, this manual implements In the example, the RGB image and the depth image corresponding to the RGB image can be used to avoid the above-mentioned problem of missed detection.
- step 108 face recognition is performed based on the target face.
- “Face brush payment” is a payment method based on face recognition, which has become one of the main payment methods for offline consumption scenarios. It has the characteristics of convenient operation and good experience. With the development of face recognition technology, “brush payment” can complete the payment behavior without requiring the user to input other identity information (such as mobile phone number and account number), that is, the user can directly complete the payment behavior only by swiping his face. For the above face brushing process, it has a risk problem: when there are multiple faces in the picture used for face brushing, it is difficult to confirm which user in the picture is willing to pay, and at this time, errors may occur. In the case of deducting money, if this happens, capital loss will occur, which will have a greater impact on the completeness of "brush payment”.
- an RGB image and a corresponding depth image for “brush payment” may be acquired, a human face in the RGB image may be detected, and a possible payment user face (that is, a target face) may be selected; thereafter, according to Select the face and depth image to determine whether there are disturbing faces in the RGB image.
- a depth image corresponding to the RGB image may be combined to determine a face for face recognition in the RGB image. .
- the depth image contains rich information, and the depth image can reflect the distance from each face in the depth image to the image acquisition device, and the face The distance to the image acquisition device can reflect the user's face recognition willingness to a certain extent. Therefore, the embodiments of this specification can avoid missing detection of the face in the RGB image, and can more accurately determine the use of the face in the RGB image. Recognized faces.
- FIG. 2 is a flowchart of a face recognition method according to another embodiment of the present specification.
- the distance from the target face to the image acquisition device may be calculated first, and the RGB may be determined based on the calculated distance and depth image. Whether there are disturbing faces in the image.
- the method may include the following steps:
- step 202 an RGB image and a corresponding depth image for face recognition are acquired, where the RGB image includes at least one face.
- the RGB image (color map) and corresponding depth image used for face recognition are images taken for the same scene.
- the gray value of each pixel in the depth image can be used to characterize the distance from a point in the shooting scene to the depth image acquisition device.
- the device used to capture depth images is called a depth image capture device, and the device used to capture RGB color images is called an RGB image capture device.
- a target face is selected from the RGB image.
- the target face image is the face most likely to be used for face recognition in the RGB image.
- face detection may be performed on an RGB image, a face contained therein is detected, and a face is selected from the face as a target face.
- a human face in a preset area in the RGB image may be selected as a target human face.
- the preset area may include: the central area of the RGB image, or Focus area when shooting RGB images.
- the face in the center area of the RGB image can be selected as the target face; or the face in the focus area when the RGB image is taken can be selected as the target face.
- step 206 a target region corresponding to the target face in the depth image is determined.
- the camera of the RGB image acquisition device and the camera of the depth image acquisition device are pre-calibrated, that is, the two have a clear spatial coordinate transformation relationship.
- the RGB image and its corresponding The spatial coordinate transformation relationship of the depth image is used to determine the coordinates (that is, the target area) of the target face on the depth image.
- step 208 the distance D1 from the target face to the face image acquisition device is calculated according to the information of the pixels in the target area.
- the distance D1 from the target face to the face image acquisition device can be calculated based on the information of the pixel points in the target area; specifically, the distance within the target area can be calculated.
- the distance from each pixel to the face image acquisition device is an average value of the distance from each pixel to the face image acquisition device to be determined as the distance D1 from the target face to the face image acquisition device.
- step 210 it is determined whether there is a face in the depth image whose distance is from the face image acquisition device is D2; if not, step 212 is performed; wherein the difference between D1 and D2 is less than a preset threshold.
- the face image acquisition device refers to a depth image acquisition device.
- the distance between the interference face and the target face to the depth image acquisition device is equal or similar.
- a face whose distance image collection device is D2 in the depth image includes a face with a complete and clear outline or a face with an incomplete and unclear outline.
- RGB image To determine whether there are disturbing faces in the RGB image, To determine whether the target face is the face with the most face recognition intent in a multi-person scene; specifically, if there is an interference face in the RGB image, it indicates that the target face is not the face with the most face-recognition intent in a multi-person scene. Human face; if there is no disturbing face in the RGB image, it indicates that the target face is the face with the most intent of face recognition in the multi-person scene.
- the face detection of the RGB image may sometimes cause the face to be missed, for example, the face in the corner of the RGB image or the half face appearing in the RGB image cannot be detected.
- the The RGB image and the depth image corresponding to the RGB image can avoid the above-mentioned problem of missed detection.
- step 212 face recognition is performed based on the target face.
- a depth image corresponding to the RGB image may be combined to determine a face for face recognition in the RGB image. .
- the depth image contains rich information, and the depth image can reflect the distance from each face in the depth image to the image acquisition device, and the face The distance to the image acquisition device can reflect the user's face recognition willingness to a certain extent. Therefore, the embodiments of this specification can avoid missing detection of the face in the RGB image, and can more accurately determine the use of the face in the RGB image. Recognized faces.
- FIG. 3 is a schematic structural diagram of a face recognition device according to an embodiment of the present specification.
- the face recognition device 300 may include: an acquisition module 301, a selection module 302, and a judgment.
- Module 303 and identification module 304 where:
- An obtaining module 301 configured to obtain an RGB image and a corresponding depth image for face recognition, where the RGB image includes at least one face;
- a selection module 302 configured to select a target face from the RGB image
- a judging module 303 is configured to judge whether an interference face exists in the RGB image according to the target face and the depth image, and a distance from the interference face to a face image acquisition device is equal to that of the target face. The difference between the distances of the face image acquisition devices is less than a preset threshold;
- the recognition module 304 is configured to perform face recognition based on the target face when the interference face does not exist in the RGB image.
- a depth image corresponding to the RGB image may be combined to determine a face for face recognition in the RGB image. .
- the depth image contains rich information, and the depth image can reflect the distance from each face in the depth image to the image acquisition device, and the face The distance to the image acquisition device can reflect the user's face recognition willingness to a certain extent. Therefore, the embodiments of this specification can avoid missing detection of the face in the RGB image, and can more accurately determine the use of the face in the RGB image. Recognized faces.
- the selection module 302 may include:
- a face selection sub-module is configured to select a face in a preset area in the RGB image as a target face.
- the preset area includes:
- a center area of the RGB image, or a focus area during shooting of the RGB image is a center area of the RGB image, or a focus area during shooting of the RGB image.
- the determining module 303 may include:
- a target region determination submodule configured to determine a target region corresponding to the target face in the depth image
- a distance calculation sub-module configured to calculate a distance D1 from the target face to a face image acquisition device according to information about pixels in the target area;
- a judging sub-module configured to judge whether there is a human face in the depth image that is D2 from the face image acquisition device, and the difference between D1 and D2 is less than the preset threshold;
- the face image acquisition device If there is a face in the depth image that is D2 from the face image acquisition device, there is an interfering face in the RGB image; if there is no distance in the depth image, the face image acquisition device is D2 Face, it is determined that there are no disturbing faces in the RGB image.
- the distance calculation submodule may include:
- a distance calculation unit configured to calculate a distance from each pixel in the target area to a face image acquisition device
- the distance determining unit is configured to determine an average value of the distances between the pixels and the face image acquisition device as the distance D1 from the target face to the face image acquisition device.
- the face recognition device 300 may further include:
- An output module is configured to output a prompt message when the interference face exists in the RGB image, and the prompt message is used to prompt the interference face in the RGB image.
- FIG. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present specification.
- the electronic device includes a processor and optionally an internal bus, a network interface, and a memory.
- the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
- RAM Random-Access Memory
- non-volatile memory such as at least one disk memory.
- the electronic device may also include hardware required for other services.
- the processor, network interface and memory can be connected to each other through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture (Extended Industry Standard Architecture) bus and so on.
- the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only a two-way arrow is used in FIG. 4, but it does not mean that there is only one bus or one type of bus.
- the program may include program code, where the program code includes a computer operation instruction.
- the memory may include memory and non-volatile memory, and provide instructions and data to the processor.
- the processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to form a face recognition device on a logical level.
- the processor executes a program stored in the memory, and is specifically used to perform the following operations:
- RGB image and a corresponding depth image for face recognition, the RGB image including at least one face;
- a depth image corresponding to the RGB image may be combined to determine a human face in the RGB image for face recognition.
- the depth image contains rich information, and the depth image can reflect the distance from each face in the depth image to the image acquisition device, and the face The distance to the image acquisition device can reflect the user's face recognition willingness to a certain extent. Therefore, the embodiments of this specification can avoid missing detection of the face in the RGB image, and can more accurately determine the use of the face in the RGB image. Recognized faces.
- the selecting a target face from the RGB image includes:
- a human face in a preset area in the RGB image is selected as a target human face.
- the preset area includes:
- a center area of the RGB image, or a focus area during shooting of the RGB image is a center area of the RGB image, or a focus area during shooting of the RGB image.
- the determining whether an interference face exists in the RGB image according to the target face and the depth image includes:
- the face image acquisition device If there is a face in the depth image that is D2 from the face image acquisition device, there is an interfering face in the RGB image; if there is no distance in the depth image, the face image acquisition device is D2 Face, it is determined that there are no disturbing faces in the RGB image.
- the calculating a distance D1 from the target face to a face image acquisition device based on information about pixels in the target area includes:
- An average value of the distances between the pixels to the face image acquisition device is determined as the distance D1 from the target face to the face image acquisition device.
- the method further includes:
- a prompt message is output, and the prompt message is used to prompt that the interfering human face exists in the RGB image.
- the method performed by the face recognition device disclosed in the embodiment shown in FIG. 4 of the present specification may be applied to a processor, or implemented by a processor.
- the processor may be an integrated circuit chip with signal processing capabilities.
- each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software.
- the aforementioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc .; it may also be a digital signal processor (DSP), special integration Circuit (Application Specific Integrated Circuit, ASIC), Field Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
- a general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
- the steps of the method disclosed in combination with the embodiments of the present specification may be directly embodied as being executed by a hardware decoding processor, or may be executed and completed by using a combination of hardware and software modules in the decoding processor.
- the software module may be located in a mature storage medium such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, and the like.
- the storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the foregoing method in combination with its hardware.
- the electronic device can also execute the method in FIG. 1 and implement the functions of the face recognition device in the embodiment shown in FIG. 1, which will not be described in detail in the embodiments of the present specification.
- An embodiment of the present specification also provides a computer-readable storage medium, where the computer-readable storage medium stores one or more programs, the one or more programs include instructions, and the instructions should be a portable electronic device including a plurality of application programs.
- the portable electronic device can be caused to execute the method in the embodiment shown in FIG. 1, and is specifically configured to execute the following method:
- RGB image Acquiring an RGB image and a corresponding depth image for face recognition, where the RGB image includes at least one face;
- the system, device, module, or unit described in the foregoing embodiments may be specifically implemented by a computer chip or entity, or a product with a certain function.
- a typical implementation device is a computer.
- the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or A combination of any of these devices.
- Computer-readable media includes both permanent and non-persistent, removable and non-removable media.
- Information can be stored by any method or technology.
- Information may be computer-readable instructions, data structures, modules of a program, or other data.
- Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), and read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, read-only disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media may be used to store information that can be accessed by computing devices.
- computer-readable media does not include temporary computer-readable media, such as modulated data signals and carrier waves.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Collating Specific Patterns (AREA)
Abstract
本说明书实施例提供一种人脸识别方法及装置,该方法包括:获取用于人脸识别的RGB图像和对应的深度图像;从RGB图像中选择目标人脸;根据目标人脸和深度图像判断RGB图像中是否存在干扰人脸;若不存在,则基于目标人脸进行人脸识别。本说明书实施例中,在对包含多个人脸的RGB图像进行人脸识别时,可以结合对应的深度图像来确定RGB图像中用于人脸识别的人脸。由于深度图像中包含的信息比较丰富、且深度图像可以反映该深度图像中的各人脸到图像采集设备的距离、且人脸到图像采集设备的距离可以从一定程度上反映用户的人脸识别意愿,因此本说明书实施例可以避免RGB图像中人脸的漏检以及准确地确定出RGB图像中用于人脸识别的人脸。
Description
本申请涉及计算机技术领域,尤其涉及一种人脸识别方法及装置。
近年来,随着人脸识别技术的发展,“刷脸”可以应用的场景越来越多,例如刷脸支付、刷脸打卡签到、刷脸解锁门禁、刷脸认证办事等,具有操作方便、快捷等特点。但是,当用于刷脸的RGB图像中存在多个人脸时,难以确定对该RGB图像中的哪个人脸进行识别,进而导致识别失败或识别错误给用户带来损失,因此,需要提出一种人脸识别方法。
发明内容
本说明书实施例的目的是提供一种人脸识别方法及装置,本说明书实施例是这样实现的:
第一方面,提供了一种人脸识别方法,所述方法包括:
获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;
从所述RGB图像中选择目标人脸;
根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
第二方面,提供了一种人脸识别装置,所述装置包括:
获取模块,用于获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;
选择模块,用于从所述RGB图像中选择目标人脸;
判断模块,用于根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存 在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
识别模块,用于在所述RGB图像中不存在所述干扰人脸的情况下,基于所述目标人脸进行人脸识别。
第三方面,提供了一种电子设备,包括:
处理器;以及
被安排成存储计算机可执行指令的存储器,所述可执行指令在被执行时使所述处理器执行以下操作:
获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;
从所述RGB图像中选择目标人脸;
根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
第四方面,提供了一种计算机存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下操作:
获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;
从所述RGB图像中选择目标人脸;
根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
由以上本说明书实施例提供的技术方案可见,本说明书实施例中,在对包含多个人脸的RGB图像进行人脸识别时,可以结合该RGB图像对应的深度图像,来确定该RGB 图像中用于人脸识别的人脸。相对于仅仅依据RGB图像进行人脸识别,本说明书实施例中,由于深度图像中包含的信息比较丰富、且深度图像可以反映该深度图像中的各人脸到图像采集设备的距离、且人脸到图像采集设备的距离可以从一定程度上反映了用户的人脸识别意愿,因此本说明书实施例可以避免RGB图像中人脸的漏检,以及可以比较准确地确定出RGB图像中用于人脸识别的人脸。
为了更清楚地说明本说明书实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本说明书中记载的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是本说明书的一个实施例的人脸识别方法的流程图;
图2是本说明书的另一个实施例的人脸识别方法的流程图;
图3是本说明书的一个实施例的人脸识别装置的结构示意图;
图4是本说明书的一个实施例的电子设备的结构示意图。
为了使本技术领域的人员更好地理解本说明书中的技术方案,下面将结合本说明书实施例中的附图,对本说明书实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本说明书一部分实施例,而不是全部的实施例。基于本说明书中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都应当属于本说明书保护的范围。
本说明书实施例提供了一种人脸识别方法及装置。
下面首先对本说明书实施例提供的一种人脸识别方法进行介绍。
需要说明的是,本说明书实施例提供的人脸识别方法适用于电子设备,在实际应用中,该电子设备可以为服务器,或者,该电子设备也可以为手机、平板电脑、个人数字助理等终端设备,或者,该电子设备也可以为笔记本电脑、台式电脑、桌面机等计算机设备,本说明书实施例对此不作限定。
图1是本说明书的一个实施例的人脸识别方法的流程图,如图1所示,该方法可以包括以下步骤:步骤102、步骤104、步骤106和步骤108,其中,
在步骤102中,获取用于人脸识别的RGB图像和对应的深度图像,其中,RGB图像中包含至少一个人脸。
本说明书实施例中,用于人脸识别的RGB图像(彩色图)和对应的深度图像为针对同一场景拍摄的图像。深度图像中每个像素点的灰度值可用于表征拍摄场景中某一点到深度图像采集设备的距离。用于采集深度图像的设备称为深度图像采集设备,用于采集RGB彩色图像的设备称为RGB图像采集设备。
在步骤104中,从RGB图像中选择目标人脸。
本说明书实施例中,目标人脸像为RGB图中最有可能用于人脸识别的人脸。
本说明书实施例中,可以对RGB图像进行人脸检测,检测其中包含的人脸,并从中选择一个人脸,作为目标人脸。具体的,可以将RGB图像中预设区域的人脸,选择为目标人脸。
考虑到具有人脸识别意图的用户通常会正对图像采集设备的拍摄焦点或处于人群的正中位置,基于这种情况,本说明书实施例中,预设区域可以包括:RGB图像的中心区域、或者RGB图像拍摄时的焦点区域。相应的,可以将RGB图像中心区域内的人脸,选择为目标人脸;或者,可以将RGB图像拍摄时的焦点区域内的人脸,选择为目标人脸。
在步骤106中,根据目标人脸和深度图像,判断RGB图像中是否存在干扰人脸;若否,则执行步骤108;其中,干扰人脸到人脸图像采集设备的距离与目标人脸到人脸图像采集设备的距离的差值小于预设阈值。
本说明书实施例中,人脸图像采集设备指的是深度图像采集设备。干扰人脸与目标人脸到深度图像采集设备的距离相当或相差不多。
考虑到具有人脸识别意图的用户通常比较靠近图像采集设备、并且在多人场景下具有人脸识别意图的用户通常只有一个,基于这种情况,本说明书实施例中,通过判断RGB图像中是否有干扰人脸,来确定目标人脸是否为多人场景下最具人脸识别意图的人脸;具体的,如果RGB图像中存在干扰人脸,则表明目标人脸不是多人场景下最具人脸识别意图的人脸;如果RGB图像中不存在干扰人脸,则表明目标人脸时多人场景下最具人脸识别意图的人脸。
考虑到对RGB图像进行人脸检测,有时会造成人脸的漏检,例如RGB图像的角落里的人脸或RGB图像中出现的半张人脸无法检测出来,基于这种情况,本说明书实施例中,采用对RGB图像和该RGB图像对应的深度图像,可以避免上述漏检的问题。
在步骤108中,基于目标人脸进行人脸识别。
本说明书实施例中,如果RGB图像中不存在干扰人脸,则基于RGB图像中的目标人脸进行人脸识别;如果RGB图像中存在干扰人脸,则输出提示消息,该提示消息用于提示RGB图像中存在干扰人脸。
为了便于理解,以“刷脸支付”场景为例对本说明书实施例的技术方案进行举例说明。
“刷脸支付”是基于人脸识别的支付方式,已成为线下消费场景的主要支付手段之一,具有操作便捷、体验好等特点。随着人脸识别技术的发展,“刷脸支付”已无需用户输入其他身份信息(如手机号、账号)便可完成支付行为,即仅需用户刷一下脸就可以直接完成支付行为。对于以上的刷脸流程而言,其具有一个风险问题:当用于刷脸的画面中存在多个人脸时,难以确认该画面中的哪个用户有意愿进行支付行为,此时,可能会出现误扣钱的情况,如果发生该情况,会发生资损,对“刷脸支付”的完全性造成较大的影响。
考虑到随着摄像头硬件的逐步发展,线下支付场景中通常都配备了深度图像采集设备,而深度图像采集设备所采集到的深度图像可以表示每个物体到相机的距离,基于这种情况,本说明书实施例中,可以获取用于“刷脸支付”的RGB图像和对应的深度图像,检测RGB图像中的人脸,选定可能的支付用户人脸(即目标人脸);之后,根据选定的人脸和深度图像,判断RGB图像中是否存在干扰人脸,如果RGB图像中存在干扰人脸,则认为本次支付交易存在多人脸无法确认的风险,并提示用户该风险,让用户再次输入相关账户信息进行确认;如果RGB图像中不存在干扰人脸,则认为本次支付交易较为安全,基于选定的人脸进行识别,识别通过后进行支付。
由上述实施例可见,该实施例中,在对包含多个人脸的RGB图像进行人脸识别时,可以结合该RGB图像对应的深度图像,来确定该RGB图像中用于人脸识别的人脸。相对于仅仅依据RGB图像进行人脸识别,本说明书实施例中,由于深度图像中包含的信息比较丰富、且深度图像可以反映该深度图像中的各人脸到图像采集设备的距离、且人脸到图像采集设备的距离可以从一定程度上反映了用户的人脸识别意愿,因此本说明书 实施例可以避免RGB图像中人脸的漏检,以及可以比较准确地确定出RGB图像中用于人脸识别的人脸。
图2是本说明书的另一个实施例的人脸识别方法的流程图,本说明书实施例中,可以首先计算目标人脸到图像采集设备的距离,根据计算得到的距离和深度图像,来判断RGB图像中是否存在干扰人脸,此时,如图2所示,该方法可以包括以下步骤:
在步骤202中,获取用于人脸识别的RGB图像和对应的深度图像,其中,RGB图像中包含至少一个人脸。
本说明书实施例中,用于人脸识别的RGB图像(彩色图)和对应的深度图像为针对同一场景拍摄的图像。深度图像中每个像素点的灰度值可用于表征拍摄场景中某一点到深度图像采集设备的距离。用于采集深度图像的设备称为深度图像采集设备,用于采集RGB彩色图像的设备称为RGB图像采集设备。
在步骤204中,从RGB图像中选择目标人脸。
本说明书实施例中,目标人脸像为RGB图中最有可能用于人脸识别的人脸。
本说明书实施例中,可以对RGB图像进行人脸检测,检测其中包含的人脸,并从中选择一个人脸,作为目标人脸。具体的,可以将RGB图像中预设区域的人脸,选择为目标人脸。
考虑到具有人脸识别意图的用户通常会正对图像采集设备的拍摄焦点或处于人群的正中位置,基于这种情况,本说明书实施例中,预设区域可以包括:RGB图像的中心区域、或者RGB图像拍摄时的焦点区域。相应的,可以将RGB图像中心区域内的人脸,选择为目标人脸;或者,可以将RGB图像拍摄时的焦点区域内的人脸,选择为目标人脸。
在步骤206中,确定目标人脸在深度图像中对应的目标区域。
考虑到RGB图像采集设备的摄像头和深度图像采集设备的摄像头是预先标定好的,即两者具有明确的空间坐标变换关系,基于这种情况,本说明书实施例中,可以根据RGB图像和其对应的深度图像的空间坐标变换关系,确定目标人脸在深度图像上的坐标(即目标区域)。
在步骤208中,根据目标区域内像素点的信息,计算目标人脸到人脸图像采集设备的距离D1。
由于深度图像中每个像素都表示距离,因此本说明书实施例中,可以根据目标区域内像素点的信息,计算目标人脸到人脸图像采集设备的距离D1;具体的,可以计算目标区域内各像素点到人脸图像采集设备的距离,将各像素点到人脸图像采集设备的距离的平均值,确定为目标人脸到人脸图像采集设备的距离D1。
在步骤210中,判断深度图像中是否存在距离人脸图像采集设备为D2的人脸;若否,则执行步骤212;其中,D1与D2的差值小于预设阈值。
本说明书实施例中,如果深度图像中存在距离人脸图像采集设备为D2的人脸,则RGB图像中存在干扰人脸;如果深度图像中不存在距离人脸图像采集设备为D2的人脸,则确定RGB图像中不存在干扰人脸。
本说明书实施例中,人脸图像采集设备指的是深度图像采集设备。干扰人脸与目标人脸到深度图像采集设备的距离相当或相差不多。
本说明书实施例中,深度图像中距离人脸图像采集设备为D2的人脸包括:轮廓完整清晰的人脸、或者轮廓不完整不清晰的人脸。
考虑到具有人脸识别意图的用户通常比较靠近图像采集设备、并且在多人场景下具有人脸识别意图的用户通常只有一个,本说明书实施例中,通过判断RGB图像中是否有干扰人脸,来确定目标人脸是否为多人场景下最具人脸识别意图的人脸;具体的,如果RGB图像中存在干扰人脸,则表明目标人脸不是多人场景下最具人脸识别意图的人脸;如果RGB图像中不存在干扰人脸,则表明目标人脸时多人场景下最具人脸识别意图的人脸。
考虑到对RGB图像进行人脸检测,有时会造成人脸的漏检,例如RGB图像的角落里的人脸或RGB图像中出现的半张人脸无法检测出来,本说明书实施例中,采用对RGB图像和该RGB图像对应的深度图像,可以避免上述漏检的问题。
在步骤212中,基于目标人脸进行人脸识别。
本说明书实施例中,如果RGB图像中不存在干扰人脸,则基于RGB图像中的目标人脸进行人脸识别;如果RGB图像中存在干扰人脸,则输出提示消息,该提示消息用于提示RGB图像中存在干扰人脸。
由上述实施例可见,该实施例中,在对包含多个人脸的RGB图像进行人脸识别时,可以结合该RGB图像对应的深度图像,来确定该RGB图像中用于人脸识别的人脸。相对于仅仅依据RGB图像进行人脸识别,本说明书实施例中,由于深度图像中包含的信 息比较丰富、且深度图像可以反映该深度图像中的各人脸到图像采集设备的距离、且人脸到图像采集设备的距离可以从一定程度上反映了用户的人脸识别意愿,因此本说明书实施例可以避免RGB图像中人脸的漏检,以及可以比较准确地确定出RGB图像中用于人脸识别的人脸。
图3是本说明书的一个实施例的人脸识别装置的结构示意图,如图3所示,在一种软件实施方式中,人脸识别装置300,可以包括:获取模块301、选择模块302、判断模块303和识别模块304,其中,
获取模块301,用于获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;
选择模块302,用于从所述RGB图像中选择目标人脸;
判断模块303,用于根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
识别模块304,用于在所述RGB图像中不存在所述干扰人脸的情况下,基于所述目标人脸进行人脸识别。
由上述实施例可见,该实施例中,在对包含多个人脸的RGB图像进行人脸识别时,可以结合该RGB图像对应的深度图像,来确定该RGB图像中用于人脸识别的人脸。相对于仅仅依据RGB图像进行人脸识别,本说明书实施例中,由于深度图像中包含的信息比较丰富、且深度图像可以反映该深度图像中的各人脸到图像采集设备的距离、且人脸到图像采集设备的距离可以从一定程度上反映了用户的人脸识别意愿,因此本说明书实施例可以避免RGB图像中人脸的漏检,以及可以比较准确地确定出RGB图像中用于人脸识别的人脸。
可选地,作为一个实施例,所述选择模块302,可以包括:
人脸选择子模块,用于将所述RGB图像中预设区域的人脸,选择为目标人脸。
可选地,作为一个实施例,所述预设区域包括:
所述RGB图像的中心区域、或者所述RGB图像拍摄时的焦点区域。
可选地,作为一个实施例,所述判断模块303,可以包括:
目标区域确定子模块,用于确定所述目标人脸在所述深度图像中对应的目标区域;
距离计算子模块,用于根据所述目标区域内像素点的信息,计算所述目标人脸到人脸图像采集设备的距离D1;
判断子模块,用于判断所述深度图像中是否存在距离所述人脸图像采集设备为D2的人脸,所述D1与D2的差值小于所述预设阈值;其中,
如果所述深度图像中存在距离所述人脸图像采集设备为D2的人脸,则所述RGB图像中存在干扰人脸;如果所述深度图像中不存在距离所述人脸图像采集设备为D2的人脸,则确定所述RGB图像中不存在干扰人脸。
可选地,作为一个实施例,所述距离计算子模块,可以包括:
距离计算单元,用于计算所述目标区域内各像素点到人脸图像采集设备的距离;
距离确定单元,用于将所述各像素点到人脸图像采集设备的距离的平均值,确定为所述目标人脸到人脸图像采集设备的距离D1。
可选地,作为一个实施例,所述人脸识别装置300,还可以包括:
输出模块,用于在所述RGB图像中存在所述干扰人脸的情况下,输出提示消息,所述提示消息用于提示所述RGB图像中存在干扰人脸。
图4是本说明书的一个实施例电子设备的结构示意图,如图4所示,在硬件层面,该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
处理器、网络接口和存储器可以通过内部总线相互连接,该内部总线可以是ISA(Industry Standard Architecture,工业标准体系结构)总线、PCI(Peripheral Component Interconnect,外设部件互连标准)总线或EISA(Extended Industry Standard Architecture,扩展工业标准结构)总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示,图4中仅用一个双向箭头表示,但并不表示仅有一根总线或一种类型的总线。
存储器,用于存放程序。具体地,程序可以包括程序代码,所述程序代码包括计算机操作指令。存储器可以包括内存和非易失性存储器,并向处理器提供指令和数据。
处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成人脸识别装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;
从所述RGB图像中选择目标人脸;
根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
本说明书实施例中,在对包含多个人脸的RGB图像进行人脸识别时,可以结合该RGB图像对应的深度图像,来确定该RGB图像中用于人脸识别的人脸。相对于仅仅依据RGB图像进行人脸识别,本说明书实施例中,由于深度图像中包含的信息比较丰富、且深度图像可以反映该深度图像中的各人脸到图像采集设备的距离、且人脸到图像采集设备的距离可以从一定程度上反映了用户的人脸识别意愿,因此本说明书实施例可以避免RGB图像中人脸的漏检,以及可以比较准确地确定出RGB图像中用于人脸识别的人脸。
可选地,作为一个实施例,所述从所述RGB图像中选择目标人脸,包括:
将所述RGB图像中预设区域的人脸,选择为目标人脸。
可选地,作为一个实施例,所述预设区域包括:
所述RGB图像的中心区域、或者所述RGB图像拍摄时的焦点区域。
可选地,作为一个实施例,所述根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,包括:
确定所述目标人脸在所述深度图像中对应的目标区域;
根据所述目标区域内像素点的信息,计算所述目标人脸到人脸图像采集设备的距离D1;
判断所述深度图像中是否存在距离所述人脸图像采集设备为D2的人脸,所述D1与D2的差值小于所述预设阈值;
如果所述深度图像中存在距离所述人脸图像采集设备为D2的人脸,则所述RGB图像中存在干扰人脸;如果所述深度图像中不存在距离所述人脸图像采集设备为D2的人脸,则确定所述RGB图像中不存在干扰人脸。
可选地,作为一个实施例,所述根据所述目标区域内像素点的信息,计算所述目标人脸到人脸图像采集设备的距离D1,包括:
计算所述目标区域内各像素点到人脸图像采集设备的距离;
将所述各像素点到人脸图像采集设备的距离的平均值,确定为所述目标人脸到人脸图像采集设备的距离D1。
可选地,作为一个实施例,所述方法还包括:
如果所述RGB图像中存在所述干扰人脸,则输出提示消息,所述提示消息用于提示所述RGB图像中存在干扰人脸。
上述如本说明书图4所示实施例揭示的人脸识别装置执行的方法可以应用于处理器中,或者由处理器实现。处理器可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器,包括中央处理器(Central Processing Unit,CPU)、网络处理器(Network Processor,NP)等;还可以是数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本说明书实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本说明书实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。
该电子设备还可执行图1的方法,并实现人脸识别装置在图1所示实施例的功能,本说明书实施例在此不再赘述。
本说明书实施例还提供了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图1所示实施例的方法,并具体用于执行以下方法:
获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少 一个人脸;
从所述RGB图像中选择目标人脸;
根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;
如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
总之,以上所述仅为本说明书的较佳实施例而已,并非用于限定本说明书的保护范围。凡在本说明书的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本说明书的保护范围之内。
上述实施例阐明的系统、装置、模块或单元,具体可以由计算机芯片或实体实现,或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的,计算机例如可以为个人计算机、膝上型计算机、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的 部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
Claims (10)
- 一种人脸识别方法,所述方法包括:获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;从所述RGB图像中选择目标人脸;根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
- 根据权利要求1所述的方法,所述从所述RGB图像中选择目标人脸,包括:将所述RGB图像中预设区域的人脸,选择为目标人脸。
- 根据权利要求2所述的方法,所述预设区域包括:所述RGB图像的中心区域、或者所述RGB图像拍摄时的焦点区域。
- 根据权利要求1所述的方法,所述根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,包括:确定所述目标人脸在所述深度图像中对应的目标区域;根据所述目标区域内像素点的信息,计算所述目标人脸到人脸图像采集设备的距离D1;判断所述深度图像中是否存在距离所述人脸图像采集设备为D2的人脸,所述D1与D2的差值小于所述预设阈值;如果所述深度图像中存在距离所述人脸图像采集设备为D2的人脸,则所述RGB图像中存在干扰人脸;如果所述深度图像中不存在距离所述人脸图像采集设备为D2的人脸,则确定所述RGB图像中不存在干扰人脸。
- 根据权利要求1所述的方法,所述根据所述目标区域内像素点的信息,计算所述目标人脸到人脸图像采集设备的距离D1,包括:计算所述目标区域内各像素点到人脸图像采集设备的距离;将所述各像素点到人脸图像采集设备的距离的平均值,确定为所述目标人脸到人脸图像采集设备的距离D1。
- 根据权利要求1所述的方法,所述方法还包括:如果所述RGB图像中存在所述干扰人脸,则输出提示消息,所述提示消息用于提示所述RGB图像中存在干扰人脸。
- 一种人脸识别装置,所述装置包括:获取模块,用于获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;选择模块,用于从所述RGB图像中选择目标人脸;判断模块,用于根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;识别模块,用于在所述RGB图像中不存在所述干扰人脸的情况下,基于所述目标人脸进行人脸识别。
- 根据权利要求7所述的装置,所述选择模块,包括:人脸选择子模块,用于将所述RGB图像中预设区域的人脸,选择为目标人脸。
- 一种电子设备,包括:处理器;以及被安排成存储计算机可执行指令的存储器,所述可执行指令在被执行时使所述处理器执行以下操作:获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;从所述RGB图像中选择目标人脸;根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
- 一种计算机存储介质,所述计算机可读存储介质存储一个或多个程序,所述一个或多个程序当被包括多个应用程序的电子设备执行时,使得所述电子设备执行以下操作:获取用于人脸识别的RGB图像和对应的深度图像,所述RGB图像中包含至少一个人脸;从所述RGB图像中选择目标人脸;根据所述目标人脸和所述深度图像,判断所述RGB图像中是否存在干扰人脸,所述干扰人脸到人脸图像采集设备的距离与所述目标人脸到所述人脸图像采集设备的距离的差值小于预设阈值;如果所述RGB图像中不存在所述干扰人脸,则基于所述目标人脸进行人脸识别。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810972167.2 | 2018-08-24 | ||
CN201810972167.2A CN109190539B (zh) | 2018-08-24 | 2018-08-24 | 人脸识别方法及装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020038140A1 true WO2020038140A1 (zh) | 2020-02-27 |
Family
ID=64919537
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/095338 WO2020038140A1 (zh) | 2018-08-24 | 2019-07-10 | 人脸识别方法及装置 |
Country Status (3)
Country | Link |
---|---|
CN (2) | CN109190539B (zh) |
TW (1) | TWI716008B (zh) |
WO (1) | WO2020038140A1 (zh) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111414858A (zh) * | 2020-03-19 | 2020-07-14 | 北京迈格威科技有限公司 | 人脸识别方法、目标图像的确定方法、装置和电子系统 |
CN112733650A (zh) * | 2020-12-29 | 2021-04-30 | 深圳云天励飞技术股份有限公司 | 目标人脸检测方法、装置、终端设备及存储介质 |
CN112926498A (zh) * | 2021-03-20 | 2021-06-08 | 杭州知存智能科技有限公司 | 基于多通道融合和深度信息局部动态生成的活体检测方法 |
CN113409056A (zh) * | 2021-06-30 | 2021-09-17 | 深圳市商汤科技有限公司 | 支付方法、装置、本地识别设备、人脸支付系统及设备 |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190539B (zh) * | 2018-08-24 | 2020-07-07 | 阿里巴巴集团控股有限公司 | 人脸识别方法及装置 |
CN109905596A (zh) * | 2019-02-13 | 2019-06-18 | 深圳市云之梦科技有限公司 | 拍照方法、装置、计算机设备和存储介质 |
CN110930547A (zh) * | 2019-02-28 | 2020-03-27 | 上海商汤临港智能科技有限公司 | 车门解锁方法及装置、系统、车、电子设备和存储介质 |
CN111860069A (zh) * | 2019-04-30 | 2020-10-30 | 百度时代网络技术(北京)有限公司 | 图像处理方法和系统 |
CN110533426A (zh) * | 2019-08-02 | 2019-12-03 | 深圳蚂里奥技术有限公司 | 一种支付方法及系统 |
CN110991239A (zh) * | 2019-10-30 | 2020-04-10 | 珠海格力电器股份有限公司 | 一种身份验证方法、装置、设备和计算机可读存储介质 |
CN111462227A (zh) * | 2020-03-27 | 2020-07-28 | 海信集团有限公司 | 室内人员定位装置及方法 |
CN111583333A (zh) * | 2020-05-19 | 2020-08-25 | 常州节卡智能装备有限公司 | 基于视觉引导的测温方法、装置、电子设备及存储介质 |
CN112150448B (zh) * | 2020-09-28 | 2023-09-26 | 杭州海康威视数字技术股份有限公司 | 图像处理方法、装置及设备、存储介质 |
CN112580553A (zh) * | 2020-12-25 | 2021-03-30 | 深圳市商汤科技有限公司 | 开关控制方法、装置、计算机设备及存储介质 |
CN112818874A (zh) * | 2021-02-03 | 2021-05-18 | 东莞埃科思科技有限公司 | 一种图像处理方法、装置、设备及存储介质 |
CN113850165B (zh) * | 2021-09-13 | 2024-07-19 | 支付宝(杭州)信息技术有限公司 | 人脸识别方法和装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040042643A1 (en) * | 2002-08-28 | 2004-03-04 | Symtron Technology, Inc. | Instant face recognition system |
CN107657219A (zh) * | 2017-09-12 | 2018-02-02 | 广东欧珀移动通信有限公司 | 人脸检测方法及相关产品 |
CN107862274A (zh) * | 2017-10-31 | 2018-03-30 | 广东欧珀移动通信有限公司 | 美颜方法、装置、电子设备和计算机可读存储介质 |
CN109190539A (zh) * | 2018-08-24 | 2019-01-11 | 阿里巴巴集团控股有限公司 | 人脸识别方法及装置 |
Family Cites Families (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101211484A (zh) * | 2006-12-25 | 2008-07-02 | 成都三泰电子实业股份有限公司 | 防止在atm取款时密码被偷窥的方法及装置 |
CN101303724A (zh) * | 2007-05-10 | 2008-11-12 | 中国银联股份有限公司 | 一种认证授权方法及系统 |
CN101661557B (zh) * | 2009-09-22 | 2012-05-02 | 中国科学院上海应用物理研究所 | 一种基于智能卡的人脸识别系统及其方法 |
US8810712B2 (en) * | 2012-01-20 | 2014-08-19 | Htc Corporation | Camera system and auto focus method |
CN102610035B (zh) * | 2012-04-05 | 2014-02-12 | 广州广电运通金融电子股份有限公司 | 金融自助设备及其防偷窥系统和方法 |
US9712738B2 (en) * | 2012-04-17 | 2017-07-18 | E-Vision Smart Optics, Inc. | Systems, devices, and methods for managing camera focus |
KR20140046341A (ko) * | 2012-10-10 | 2014-04-18 | 노틸러스효성 주식회사 | 얼굴인식기능을 활용한 금융자동화기기 보안 시스템 및 보안 방법 |
CN103353935B (zh) * | 2013-07-19 | 2016-06-08 | 电子科技大学 | 一种用于智能家居系统的3d动态手势识别方法 |
CN103795864B (zh) * | 2014-01-29 | 2016-09-28 | 华为技术有限公司 | 移动终端前后摄像头的选择方法和移动终端 |
CN107493435B (zh) * | 2015-03-05 | 2020-06-26 | Oppo广东移动通信有限公司 | 一种拍摄方法及终端和相关介质产品 |
CN105260732A (zh) * | 2015-11-26 | 2016-01-20 | 小米科技有限责任公司 | 图片处理方法及装置 |
TW201727537A (zh) * | 2016-01-22 | 2017-08-01 | 鴻海精密工業股份有限公司 | 人臉識別系統及人臉識別方法 |
CN105791685B (zh) * | 2016-02-29 | 2019-03-12 | Oppo广东移动通信有限公司 | 控制方法、控制装置及电子装置 |
CN107426552A (zh) * | 2016-05-23 | 2017-12-01 | 中兴通讯股份有限公司 | 防炫目方法及装置、投影设备 |
CN106339673A (zh) * | 2016-08-19 | 2017-01-18 | 中山大学 | 一种基于人脸识别的atm机身份验证方法 |
TWI604332B (zh) * | 2017-03-24 | 2017-11-01 | 緯創資通股份有限公司 | 遠距離的身份辨識方法及其系統與電腦可讀取記錄媒體 |
CN107368778A (zh) * | 2017-06-02 | 2017-11-21 | 深圳奥比中光科技有限公司 | 人脸表情的捕捉方法、装置及存储装置 |
CN107703962A (zh) * | 2017-08-26 | 2018-02-16 | 上海瞬动科技有限公司合肥分公司 | 一种无人机集体照拍摄方法 |
CN107844744A (zh) * | 2017-10-09 | 2018-03-27 | 平安科技(深圳)有限公司 | 结合深度信息的人脸识别方法、装置及存储介质 |
CN107657245A (zh) * | 2017-10-16 | 2018-02-02 | 维沃移动通信有限公司 | 一种人脸识别方法和终端设备 |
CN107918726A (zh) * | 2017-10-18 | 2018-04-17 | 深圳市汉普电子技术开发有限公司 | 距离感应方法、设备及存储介质 |
CN107977650B (zh) * | 2017-12-21 | 2019-08-23 | 北京华捷艾米科技有限公司 | 人脸检测方法及装置 |
CN108230465B (zh) * | 2018-01-26 | 2020-05-15 | 深圳一卡通新技术有限公司 | 基于移动终端的闸机快速通行控制方法 |
-
2018
- 2018-08-24 CN CN201810972167.2A patent/CN109190539B/zh active Active
- 2018-08-24 CN CN202010729662.8A patent/CN111832535B/zh active Active
-
2019
- 2019-06-21 TW TW108121633A patent/TWI716008B/zh active
- 2019-07-10 WO PCT/CN2019/095338 patent/WO2020038140A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040042643A1 (en) * | 2002-08-28 | 2004-03-04 | Symtron Technology, Inc. | Instant face recognition system |
CN107657219A (zh) * | 2017-09-12 | 2018-02-02 | 广东欧珀移动通信有限公司 | 人脸检测方法及相关产品 |
CN107862274A (zh) * | 2017-10-31 | 2018-03-30 | 广东欧珀移动通信有限公司 | 美颜方法、装置、电子设备和计算机可读存储介质 |
CN109190539A (zh) * | 2018-08-24 | 2019-01-11 | 阿里巴巴集团控股有限公司 | 人脸识别方法及装置 |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111414858A (zh) * | 2020-03-19 | 2020-07-14 | 北京迈格威科技有限公司 | 人脸识别方法、目标图像的确定方法、装置和电子系统 |
CN111414858B (zh) * | 2020-03-19 | 2023-12-19 | 北京迈格威科技有限公司 | 人脸识别方法、目标图像的确定方法、装置和电子系统 |
CN112733650A (zh) * | 2020-12-29 | 2021-04-30 | 深圳云天励飞技术股份有限公司 | 目标人脸检测方法、装置、终端设备及存储介质 |
CN112733650B (zh) * | 2020-12-29 | 2024-05-07 | 深圳云天励飞技术股份有限公司 | 目标人脸检测方法、装置、终端设备及存储介质 |
CN112926498A (zh) * | 2021-03-20 | 2021-06-08 | 杭州知存智能科技有限公司 | 基于多通道融合和深度信息局部动态生成的活体检测方法 |
CN112926498B (zh) * | 2021-03-20 | 2024-05-24 | 杭州知存智能科技有限公司 | 基于多通道融合和深度信息局部动态生成的活体检测方法及装置 |
CN113409056A (zh) * | 2021-06-30 | 2021-09-17 | 深圳市商汤科技有限公司 | 支付方法、装置、本地识别设备、人脸支付系统及设备 |
Also Published As
Publication number | Publication date |
---|---|
CN109190539B (zh) | 2020-07-07 |
TWI716008B (zh) | 2021-01-11 |
TW202009785A (zh) | 2020-03-01 |
CN111832535B (zh) | 2024-09-06 |
CN109190539A (zh) | 2019-01-11 |
CN111832535A (zh) | 2020-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020038140A1 (zh) | 人脸识别方法及装置 | |
TWI686757B (zh) | 資源轉移方法、資金支付方法、裝置及電子設備 | |
US10817705B2 (en) | Method, apparatus, and system for resource transfer | |
US11210541B2 (en) | Liveness detection method, apparatus and computer-readable storage medium | |
CN109086734B (zh) | 一种对人眼图像中瞳孔图像进行定位的方法及装置 | |
US11210502B2 (en) | Comparison method and apparatus based on a plurality of face image frames and electronic device | |
CN110765958A (zh) | 基于人脸识别的渠道风险控制方法方法、装置及系统 | |
US11017557B2 (en) | Detection method and device thereof | |
US20210027045A1 (en) | Method and device for face selection, recognition and comparison | |
CN111274602B (zh) | 一种图像特征信息替换方法、装置、设备及介质 | |
CN109145827A (zh) | 视频通讯方法及装置 | |
CN109344732B (zh) | 一种鉴权以及确定鉴权方式的方法、装置及电子设备 | |
CN111860285B (zh) | 用户注册方法、装置、电子设备及存储介质 | |
CN109118506A (zh) | 一种确定人眼图像中瞳孔图像边缘点的方法及装置 | |
JP7268483B2 (ja) | 動画処理装置、動画処理方法、及びプログラム | |
CN111401883B (zh) | 一种应用功能和生物核身的开通方法、装置和电子设备 | |
CN113705457A (zh) | 基于人脸的业务处理方法和装置 | |
CN117237682A (zh) | 一种证件验证方法、装置、存储介质及电子设备 | |
CN118171252A (zh) | 身份识别方法、装置、计算机设备和存储介质 |
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: 19851553 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19851553 Country of ref document: EP Kind code of ref document: A1 |