Equipment wakeup method, device and system

Abstract

The invention provides an equipment awakening method, device and system, and the method comprises the steps: obtaining a target figure image of an environment sound source region under the condition that the environment sound is detected; performing action analysis on the target person image, and determining a body language of the target person; and based on the matching degree of the body language and the target body language, waking up the device. According to the equipment wake-up method, device and system provided by the invention, the target figure image of the environment sound source area is obtained under the condition that the environment sound is detected, so that whether the user has the wake-up intention can be accurately identified by combining the sound information and the visual information, and the problem that the wake-up precision is influenced by identifying the wake-up intention only based on the sound information is avoided.

Description

Technical Field

[0001] The present invention relates to the field of intelligent identification technology, and in particular to a device wake-up method, device and system. Background Art

[0002] Device wake-up means activating the device to switch it from standby or off state to working state, ready to receive user instructions or interact.

[0003] Currently, device wake-up usually relies on the device's built-in microphone to collect sound, and then uses the voice recognition algorithm to parse the voice command and perform the wake-up operation. However, when multiple users speak at the same time or in a noisy environment, accurate wake-up and precise recognition cannot be achieved, resulting in a poor product interaction experience. Summary of the invention

[0004] The present invention provides a device awakening method, apparatus and system to solve the defects in the prior art.

[0005] The present invention provides a device waking up method, comprising the following steps: When the ambient sound is detected, an image of a target person in the area where the ambient sound comes from is obtained; Performing action analysis on the target person image to determine the target person's body language; Based on the matching degree between the body language and the target body language, the device is awakened.

[0006] According to a device awakening method provided by the present invention, the step of performing motion analysis on the target person image to determine the target person's body language includes: Extracting multiple body features from the target person image; Based on the candidate body languages ​​to which each body feature belongs, the body language of the target person is determined.

[0007] According to a device awakening method provided by the present invention, determining the body language of the target person based on the candidate body language to which each body feature belongs includes: Count the number of corresponding body features under each candidate body language; The body language of the target person is determined based on the number of body features corresponding to each candidate body language.

[0008] According to a device awakening method provided by the present invention, the step of performing motion analysis on the target person image to determine the target person's body language includes: Performing body movement analysis on the target person image to determine the body movement language in the body language; Perform facial action analysis on the target person image to determine the facial action language in the body language.

[0009] According to a device awakening method provided by the present invention, the device is awakened based on the matching degree between the body language and the target body language, including: When the degree of matching between the facial action language and the target body language is less than a first threshold, determining the degree of matching between the body action language and the target body language; When the matching degree between the body language and the target body language is greater than or equal to a second threshold, the device is awakened.

[0010] According to a device wake-up method provided by the present invention, obtaining the target person image in the area where the ambient sound comes from includes: Using a sound pickup device to locate the source area; Based on the position of the source area, adjusting the camera to a target angle, the target angle being an angle corresponding to the frontal image of the person captured by the camera; At the target angle, the image of the target person is captured by the camera.

[0011] The present invention also provides a device waking up a device, comprising: An acquisition unit, configured to acquire an image of a target person in an area where the ambient sound is sourced, when the ambient sound is detected; An analysis unit, used to perform action analysis on the target person image to determine the target person's body language; The awakening unit is used to awaken the device based on the matching degree between the body language and the target body language.

[0012] The present invention also provides a device awakening system, comprising: a camera, a sound pickup device and the device awakening apparatus as described above, wherein the camera is used to collect the target person image, and the sound pickup device is used to collect the environmental sound.

[0013] A device wake-up system provided according to the present invention further includes a recognition unit and a control unit, wherein the recognition unit is used to perform semantic understanding of the ambient sound when the device is in a wake-up state, and the control instruction is used to generate a control instruction based on the semantic understanding result.

[0014] A device awakening system provided according to the present invention further includes a terminal interface, wherein the terminal interface is connected to a terminal so as to migrate the execution logic of the device awakening device to the terminal.

[0015] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned device wake-up methods is implemented.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the device wake-up method described in any one of the above is implemented.

[0017] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the device wake-up method described above is implemented.

[0018] The device wake-up method, apparatus and system provided by the present invention obtain the image of the target person in the source area of ​​the ambient sound when the ambient sound is detected, so as to accurately identify whether the user has the wake-up intention by combining the sound information and the visual information, avoiding the problem of affecting the wake-up accuracy by identifying the wake-up intention based only on the sound information. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a flow chart of the device awakening method provided by the present invention.

[0021] Figure 2 This is one of the flowchart diagrams of the implementation of step 120 in the device wake-up method provided by the present invention.

[0022] Figure 3 This is the second flowchart diagram of the implementation of step 120 in the device wake-up method provided by the present invention.

[0023] Figure 4 It is a flowchart of an implementation of step 110 in the device wake-up method provided by the present invention.

[0024] Figure 5 It is a structural schematic diagram of the device awakening apparatus provided by the present invention.

[0025] Figure 6 It is a structural schematic diagram of the device wake-up system provided by the present invention.

[0026] Figure 7 It is a schematic diagram of the installation of the camera and the microphone provided by the present invention.

[0027] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Currently, device wake-up usually relies on the device's built-in microphone to collect sound, and then uses a voice recognition algorithm to parse the voice command and perform the wake-up operation. This method can accurately wake up the device in a quiet environment with standard pronunciation, but it cannot achieve accurate wake-up and precise recognition when multiple users speak at the same time or in a noisy environment, resulting in a poor product interaction experience.

[0030] In view of the above situation, this embodiment provides a device wake-up method. It should be noted that the method can be applied to smart terminal devices with wake-up functions, such as smart speakers, wearable devices, smart home appliances, and in-vehicle smart devices. Figure 1 is a flow chart of the device wake-up method provided by the present invention, such as Figure 1 As shown, the method includes step 110 , step 120 and step 130 .

[0031] Step 110: When ambient sound is detected, an image of the target person in the area where the ambient sound comes from is obtained.

[0032] Here, the environmental sound can be understood as the sound generated in the surrounding environment, and the environmental sound can include various sounds such as human voice, object collision, traffic noise, etc. The environmental sound can be collected by a sound pickup device (such as a microphone).

[0033] The source area refers to the spatial location or area where the ambient sound is generated. For example, in a room, if someone is talking, multiple microphones or sensors can be used to calculate the direction and distance of the sound to identify the location of the sound source.

[0034] The target person image refers to a person image that matches the source of the environmental sound. For example, computer vision technology (such as facial recognition, posture analysis, etc.) can be used to determine whether there is a relevant target person in the source area. If so, the corresponding target person image can be collected.

[0035] It should be noted that if ambient sound is detected, it indicates that the user may have the intention to wake up. Considering that ambient sound may include noise (such as object collision, traffic noise, etc.) in addition to human voice, if the device is woken up based on ambient sound alone, the human voice in the ambient sound may be masked by the noise, resulting in the inability to accurately detect the wake-up word or voice command, affecting the wake-up accuracy.

[0036] In this regard, when the embodiment of the present invention detects ambient sound, it obtains the image of the target person in the source area of ​​the ambient sound to identify whether the user has the intention to wake up from the visual information level, thereby avoiding the problem of affecting the wake-up accuracy based only on the sound level.

[0037] In addition, considering that there may be multiple sounds simultaneously near the source area of ​​the ambient sound, in order to accurately match the target person corresponding to the ambient sound, the embodiment of the present invention first locates the source area of ​​the ambient sound, so that the target person image matching the ambient sound can be accurately captured in the source area.

[0038] Step 120: Perform action analysis on the target person's image to determine the target person's body language.

[0039] Step 130: wake up the device based on the matching degree between the body language and the target body language.

[0040] Specifically, the body language of the target person refers to the body movements, postures, facial expressions and other behaviors of the target person in expressing emotions, thoughts, attitudes or intentions in a non-verbal manner. Body language may include facial movements, body movements, etc. Optionally, facial movement analysis and body movement analysis may be performed on the target person image to determine the body language of the target person.

[0041] Target body language can be understood as body language used to trigger device wake-up. For example, specific body movements (such as an "OK" gesture can wake up the device, the distance between the user and the device is less than a threshold can wake up the device), specific facial movements (such as lip shape matching the ambient sound, eyes looking at the device for more than a preset time, etc. can wake up the device).

[0042] In addition, the matching degree between the body language and the target body language is used to characterize the similarity between the target person's body language and the target body language. The higher the similarity, the higher the probability that the user has the intention to awaken.

[0043] After waking up the device, the target person's line of sight can be tracked in the interactive area (such as the camera shooting area), and the target person's image can be acquired in real time to avoid the device entering the dormant state again due to the failure to capture the target person's image or incomplete capture of the target person's image during the target person's movement. The interactive area can be understood as the area where the user needs to interact with the smart terminal device. For example, the interactive area can be an area less than a threshold (such as 1 meter) from the smart terminal device.

[0044] The device wake-up method provided in an embodiment of the present invention obtains an image of a target person in the area where the ambient sound is sourced when ambient sound is detected, thereby accurately identifying whether the user has a wake-up intention by combining sound information and visual information, thereby avoiding the problem of affecting the wake-up accuracy by identifying the wake-up intention based only on sound information.

[0045] Based on the above embodiments, Figure 2 is one of the flowcharts of the implementation of step 120 in the device wake-up method provided by the present invention, such as Figure 2 As shown, step 120 performs action analysis on the target person image to determine the target person's body language, including: Step 121a, extracting multiple body features from the target person image; Step 122a: Determine the body language of the target person based on the candidate body languages ​​to which each body feature belongs.

[0046] Specifically, body features refer to features that can reflect the target person's body shape, posture, action, etc. Among them, body features can include body movement features (such as running, raising hands, etc.), facial movement features (such as gazing, lip shape movements, etc.).

[0047] Considering that each body language may contain multiple body features, after determining the candidate body languages ​​to which each body feature belongs, the candidate body language containing the largest number of body features can be used as the body language of the target person.

[0048] For example, when waking up the smart home system, you need to use the body language of "raising your hand to say hello" to wake up the device, and "raising your hand to say hello" corresponds to multiple body characteristics: arm angle (arms raised to a certain angle), hand position (the height of the hand relative to the body, and whether it is fully extended), head direction (the head may be slightly turned upward to indicate that it is attracting attention), facial expression (may be smiling, showing friendly intentions), shoulder state (shoulders may be relaxed, showing a natural hand-raising movement, rather than a sudden or nervous movement).

[0049] It can be seen that the embodiment of the present invention can accurately determine the body language of the target person based on multiple body features, and further can accurately wake up the device based on the body language of the target person.

[0050] Based on any of the above embodiments, step 122 determines the body language of the target person based on the candidate body languages ​​to which each body feature belongs, including: Count the number of corresponding body features under each candidate body language; Based on the number of corresponding body features under each candidate body language, the body language of the target person is determined.

[0051] Specifically, the same body feature may correspond to multiple candidate body languages. For example, the body feature "arm angle 45°" may correspond to multiple candidate body languages ​​such as "raise hand to say hello", "ask for help", "stop the action", etc. In other words, the body language of the target person cannot be determined based on a single body feature.

[0052] In this regard, the embodiment of the present invention counts the number of corresponding body features under each candidate body language. The more the number of corresponding body features under any candidate body language, the greater the probability that the body language of the target person is the corresponding candidate body language. Optionally, the candidate body language with the largest number of body features can be used as the body language of the target person.

[0053] Based on any of the above embodiments, Figure 3 FIG. 2 is a flow chart of an implementation of step 120 in the device wake-up method provided by the present invention. Figure 3 As shown, step 120 performs action analysis on the target person image to determine the target person's body language, including: Step 121b, performing body movement analysis on the target person image to determine the body movement language in the body language; Step 122b: perform facial action analysis on the target person image to determine the facial action language in the body language.

[0054] Specifically, body movement analysis is to extract the emotional state or psychological activity of a person at a specific moment by identifying body movements. Facial movement analysis is to identify specific facial expressions (such as smiling, frowning, opening the mouth, etc.) and infer the inner emotions of the person based on these expressions.

[0055] For example, by analyzing body movements, it can be determined whether the target person is in a suitable state for awakening. For example, if the target person's sitting or lying posture indicates that they are still in a resting state (such as crossed hands, head hanging down, body relaxed, etc.), the device can be awakened later to avoid being disturbed. If the target person has movements, such as slight stretching, arm or head rotation, etc., it may mean that they have begun to wake up, and the device wake-up process can be initiated.

[0056] For example, by analyzing body movements, the target person's need to be awakened can be determined. For example, if the target person stretches or turns over, the device can infer that the user is transitioning from light sleep to wakefulness through these body movements, and then send a gentle wake-up signal to avoid suddenly interrupting the target person's deep sleep. If it is detected that the target person is already in a fully awake posture (such as standing, sitting up, etc.), the device can send a wake-up prompt in a more direct way.

[0057] It can be seen that the embodiment of the present invention determines the body language in the body language by analyzing the body movements of the target person image, so that the wake-up method can be adjusted according to the body language. For example, if the target person shows anxious or unhappy body language (such as clenching his hands), the device sound can be adjusted to avoid an overly abrupt wake-up method.

[0058] For example, by analyzing facial movements, you can determine the target person's emotional state and whether they are ready to be awakened. For example, facial expressions such as frowning, closed eyes, and tightly closed lips may indicate that they are still in deep sleep or not ready to wake up, and you can choose a gentle awakening method or a slight delay. Facial expressions such as wide eyes, slight smile, and slightly opened lips may indicate that the target person is already in a semi-awake state, and the intensity of awakening can be gradually increased.

[0059] It can be seen that the embodiment of the present invention determines the facial action language in the body language by performing facial action analysis on the target person image, so that the awakening method can be adjusted according to the facial action language, avoiding an overly abrupt awakening method and improving the user experience.

[0060] Based on any of the above embodiments, waking up the device based on the matching degree between the body language and the target body language includes: When the matching degree between the facial action language and the target body language is less than a first threshold, determining the matching degree between the body action language and the target body language; When the matching degree between the body language and the target body language is greater than or equal to a second threshold, the device is awakened.

[0061] Specifically, the matching degree between facial action language and target body language is used to reflect the target person's awakening needs from the facial action level, and the matching degree between body action language and target body language is used to reflect the target person's awakening needs from the body action level.

[0062] In addition, consider that facial action language (such as the degree of eye openness, eyebrow movement, the state of the corners of the mouth, etc.) can usually quickly and accurately reflect the target person's emotional changes and state of wakefulness. Facial expression changes are more direct and subtle than body movements, and can provide key information about whether the target person is really ready to wake up. Although body movement language (such as turning over, stretching, getting up, etc.) can also provide important information about the target person's state of wakefulness, its response is usually slow or requires more obvious movements to represent. For example, the target person may show that he is ready to wake up by turning his body, stretching his limbs, sitting up, etc. In contrast, facial expressions may begin to change long before body movements.

[0063] In this regard, an embodiment of the present invention adopts the judgment order of facial movements first and then body movements, that is, it is first judged whether the matching degree between facial movement language and target body language is less than a first threshold. If so, it indicates that the user's wake-up demand cannot be determined based on facial movement language at present, and then the matching degree between body movement language and target body language is determined.

[0064] It can be seen that the embodiment of the present invention can accurately determine the best time to wake up the device by gradually screening and analyzing, combining facial action language and body action language, avoiding waking up the user too early or too late, and improving the overall user experience.

[0065] Based on any of the above embodiments, Figure 4 is a flow chart of an implementation of step 110 in the device wake-up method provided by the present invention, such as Figure 4 As shown, step 110 obtains the target person image in the ambient sound source area, including: Step 111: using a sound pickup device to locate the source area; Step 112: Based on the location of the source area, adjust the camera to a target angle, where the target angle is the angle corresponding to the frontal image of the person captured by the camera; Step 113: At the target angle, capture the image of the target person through the camera.

[0066] Specifically, the sound pickup device may be a microphone array, and the sound pickup device may use sound source localization technology to locate the source area of ​​the ambient sound. For example, the sound pickup device can calculate the propagation direction of the sound by analyzing the time difference and intensity of the sound signals received by different microphones, and then determine the source area of ​​the ambient sound.

[0067] After determining the location of the source area, adjust the camera to the target angle so that the front image of the target person can be clearly captured. The target angle refers to the position that the camera should be facing so that the camera can capture the front image of the target person, ensuring that the camera captures the target person's complete facial expressions, movements and other details.

[0068] For example, the camera may be mechanically adjustable (e.g., using a pan / tilt control or an electric rotation device), or multiple cameras may be combined into an array. By analyzing the position of the source area, the optimal adjustment angle, i.e., the target angle, is calculated.

[0069] After the camera is adjusted to the target angle, the camera starts to collect the target person's image. At this time, the camera is aimed at the front of the target person to shoot, so as to capture the target person's complete facial expression, movements and other details, and then accurately analyze the target person's image, determine the target person's body language, and accurately wake up the device based on the target person's body language.

[0070] It has been verified that the device wake-up method provided by the embodiment of the present invention has a wake-up recognition accuracy rate of >95% in a quiet and noise-free scene, and a wake-up recognition accuracy rate of >93% in a noisy scene (such as a scene with a signal-to-noise ratio of 0dB).

[0071] The device awakening apparatus provided by the present invention is described below. The device awakening apparatus described below and the device awakening method described above can be referenced to each other.

[0072] Based on any of the above embodiments, Figure 5 is a schematic diagram of the structure of the device wake-up device provided by the present invention, such as Figure 5 As shown, the device comprises: An acquisition unit 510 is used to acquire a target person image in the area where the ambient sound comes from when the ambient sound is detected; An analysis unit 520 is used to perform action analysis on the target person image to determine the target person's body language; The awakening unit 530 is used to awaken the device based on the matching degree between the body language and the target body language.

[0073] The device wake-up device provided by the present invention obtains the image of the target person in the source area of ​​the ambient sound when the ambient sound is detected, so as to accurately identify whether the user has the wake-up intention by combining the sound information and the visual information, thereby avoiding the problem of affecting the wake-up accuracy by identifying the wake-up intention based only on the sound information.

[0074] Based on any of the above embodiments, performing action analysis on the target person image to determine the target person's body language includes: Extract multiple body features from the target person image; Based on the candidate body languages ​​to which each body feature belongs, the body language of the target person is determined.

[0075] Based on any of the above embodiments, determining the body language of the target person based on the candidate body languages ​​to which each body feature belongs includes: Count the number of corresponding body features under each candidate body language; Based on the number of corresponding body features under each candidate body language, the body language of the target person is determined.

[0076] Based on any of the above embodiments, performing action analysis on the target person image to determine the target person's body language includes: Perform body movement analysis on the target person’s image to determine the body language in the body language; Perform facial action analysis on the target person's image to determine the facial action language in the body language.

[0077] Based on any of the above embodiments, waking up the device based on the matching degree between the body language and the target body language includes: When the matching degree between the facial action language and the target body language is less than a first threshold, determining the matching degree between the body action language and the target body language; When the matching degree between the body language and the target body language is greater than or equal to a second threshold, the device is awakened.

[0078] Based on any of the above embodiments, obtaining an image of a target person in an area where ambient sound comes from includes: Use sound pickup equipment to locate the source area; Based on the location of the source area, adjust the camera to the target angle, where the target angle is the angle at which the camera captures the frontal image of the person; At the target angle, the image of the target person is collected through the camera.

[0079] Based on any of the above embodiments, the present invention further provides a device wake-up system, the system comprising: A camera, a sound pickup device, and a device wake-up apparatus as described in any of the above embodiments, wherein the camera is used to capture images of the target person, and the sound pickup device is used to capture ambient sounds.

[0080] Specifically, the sound pickup device is used to collect ambient sound, and after collecting the ambient sound, the ambient sound is sent to the device wake-up device, so that the device wake-up device controls the camera to collect the target person image in the source area of ​​the ambient sound. After the camera collects the target person image, the target person image is sent to the device wake-up device, so that the device wake-up device performs motion analysis on the target person image, determines the target person's body language, and wakes up the device based on the matching degree between the body language and the target body language.

[0081] Based on any of the above embodiments, the system further includes a recognition unit and a control unit, wherein the recognition unit is used to perform semantic understanding of ambient sounds when the device is in an awake state, and the control instructions are used to generate control instructions based on the semantic understanding results.

[0082] The device wake-up system provided by the present invention obtains an image of a target person in the area where the ambient sound comes from when the ambient sound is detected, so that it can accurately identify whether the user has the wake-up intention by combining the sound information and the visual information, thereby avoiding the problem of affecting the wake-up accuracy by identifying the wake-up intention based only on the sound information.

[0083] Figure 6 is a schematic diagram of the structure of the device wake-up system provided by the present invention, such as Figure 6 As shown, the sound pickup device can be a microphone, the control unit can be a host computer, and the host is provided with a device wake-up device and an identification unit. The host can be a 3588 host or other hosts. The microphone is connected to the sound card motherboard via an audio cable, and the sound card motherboard is connected to the host via a USB audio cable. The camera is connected to the host via a USB camera harness. After the recognition unit performs semantic understanding of the ambient sound, it sends the semantic understanding to the host computer via a serial port to USB port harness. The host computer performs logical judgment, business operation, and speech synthesis based on the semantic understanding results, and generates control instructions, wherein the control instructions can be used to interact with the target person, or can be used to execute the operation of the target person carried in the ambient sound, and the embodiment of the present invention does not specifically limit this.

[0084] For example, Figure 7 Schematic diagram of the installation of the camera and microphone provided by the present invention, such as Figure 7 As shown, the microphone is a ring microphone, which is installed horizontally on the top of the head with the microphone hole facing upward. The camera needs to face the face when working, so that the front face image of the target person can be collected. In addition, the horizontal field of view of the camera is 95°.

[0085] Based on any of the above embodiments, the system further includes a terminal interface, which is connected to the terminal to migrate the execution logic of the device waking up the apparatus to the terminal.

[0086] Exemplarily, the system rebuilds a business scheduling model with multiple functions such as voice wake-up, face wake-up, gesture wake-up, and gaze wake-up. Through the terminal interface, the parameters of the business scheduling model of one or more functions can be flexibly configured according to the terminal business requirements, so that the terminal has the corresponding wake-up function. In addition, when the terminal is configured with the corresponding function, the scheduling engine can throw out events with simple interfaces to the outside through the built-in interaction rights transfer service to shield the complex scheduling logic, thereby reducing the access difficulty of manufacturers, ensuring the interaction effect of the product, and improving the device interaction experience and intelligence level. Among them, in an environment containing multiple smart devices, the interaction rights transfer service ensures that users can interact with different devices seamlessly. It can intelligently determine which device is most suitable for the current interaction needs and transfer the interaction rights to the device. For example, if a user is interacting with a smart speaker, but other devices (such as TVs and mobile phones) need to respond to the user's instructions during the process, the interaction rights transfer service will determine which device will execute the user's request.

[0087] Figure 8 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the device wake-up method, which includes: when the ambient sound is detected, obtaining the target person image in the source area of ​​the ambient sound; performing motion analysis on the target person image to determine the body language of the target person; and waking up the device based on the matching degree between the body language and the target body language.

[0088] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0089] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the device wake-up method provided by the above-mentioned methods, and the method includes: when ambient sound is detected, obtaining an image of a target person in the source area of ​​the ambient sound; performing motion analysis on the image of the target person to determine the body language of the target person; and waking up the device based on the degree of match between the body language and the target body language.

[0090] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the device wake-up method provided by the above-mentioned methods. The method includes: when ambient sound is detected, obtaining an image of a target person in the source area of ​​the ambient sound; performing motion analysis on the image of the target person to determine the body language of the target person; and waking up the device based on the degree of match between the body language and the target body language.

[0091] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0092] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A device wake-up method, characterized in that: include: When the ambient sound is detected, an image of a target person in the area where the ambient sound comes from is obtained; Performing action analysis on the target person image to determine the target person's body language; Based on the matching degree between the body language and the target body language, the device is awakened.

2. The device wake-up method according to claim 1, characterized in that: The step of performing action analysis on the target person image to determine the target person's body language includes: Extracting multiple body features from the target person image; Based on the candidate body languages ​​to which each body feature belongs, the body language of the target person is determined.

3. The device wake-up method according to claim 2, characterized in that: The determining the body language of the target person based on the candidate body languages ​​to which each body feature belongs includes: Count the number of corresponding body features under each candidate body language; The body language of the target person is determined based on the number of body features corresponding to each candidate body language.

4. The device wake-up method according to claim 1, characterized in that: The step of performing action analysis on the target person image to determine the target person's body language includes: Performing body movement analysis on the target person image to determine the body movement language in the body language; Perform facial action analysis on the target person image to determine the facial action language in the body language.

5. The device wake-up method according to claim 4, characterized in that: The waking up the device based on the matching degree between the body language and the target body language includes: When the degree of matching between the facial action language and the target body language is less than a first threshold, determining the degree of matching between the body action language and the target body language; When the matching degree between the body language and the target body language is greater than or equal to a second threshold, the device is awakened.

6. The device wake-up method according to any one of claims 1 to 5, characterized in that: The step of obtaining the target person image in the area where the ambient sound comes from includes: Using a sound pickup device to locate the source area; Based on the position of the source area, adjusting the camera to a target angle, the target angle being an angle corresponding to the frontal image of the person captured by the camera; At the target angle, the image of the target person is captured by the camera.

7. A device waking up a device, characterized in that: include: An acquisition unit, configured to acquire an image of a target person in an area where the ambient sound is sourced, when the ambient sound is detected; An analysis unit, used to perform action analysis on the target person image to determine the target person's body language; The awakening unit is used to awaken the device based on the matching degree between the body language and the target body language.

8. A device wake-up system, characterized in that: include: A camera, a sound pickup device, and a device wake-up apparatus as described in claim 7, wherein the camera is used to capture the image of the target person, and the sound pickup device is used to capture the ambient sound.

9. The device wake-up system according to claim 8, characterized in that: The system also includes a recognition unit and a control unit. The recognition unit is used to perform semantic understanding of the ambient sound when the device is in an awake state, and the control instruction is used to generate a control instruction based on the semantic understanding result.

10. The device wake-up system according to claim 8, characterized in that: The system further comprises a terminal interface, which is connected to a terminal so as to migrate the execution logic of the device waking up the apparatus to the terminal.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the device wake-up method according to any one of claims 1 to 6 is implemented.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the device wake-up method according to any one of claims 1 to 6 is implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the device wake-up method according to any one of claims 1 to 6 is implemented.

Images