Mobile phone PIN lock screen password recognition method and system based on computer vision
Through deep learning object detection algorithm and optical flow algorithm, the automatic identification of mobile phone PIN lock screen passwords is realized, solving the high cost and inefficiency problems caused by manual participation in the existing technology, and improving the universality and success rate of recognition.
Patent Information
- Application Number
- CN202210246096.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-14
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2042-03-14
AI Technical Summary
The existing computer vision-based mobile phone PIN lock screen password recognition method requires manual participation steps such as video editing and target annotation, resulting in high analysis costs and low recognition efficiency, and it is difficult to adapt to different mobile phone models and unlocking postures.
Using a deep learning-based object detection algorithm, the position of fingertips and numeric keys is automatically detected through the YOLO series detection algorithm, combined with the optical flow algorithm to track the fingertip motion, and adjust the keyboard layout using the affine mapping principle, calculate the fingertip motion parameters to judge the keys state, realizes automatic identification and object detection.
It reduces labor costs, improves the versatility and efficiency of recognition, and can automatically process different mobile phone models and unlocking postures, improving the recognition success rate.
Smart Images

Figure CN114820720B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of video analysis, and in particular relates to a method and system for identifying a mobile phone PIN lock screen password based on computer vision. Background Art
[0002] At present, mobile devices such as smartphones, wearable devices and tablets have become ubiquitous computing and communication platforms, playing an increasingly important role in daily life, and their security is becoming more and more important. PIN is an identity authentication method that takes into account both security and convenience, and is therefore the default unlocking method for most mobile phone lock screens. PIN is generally composed of fixed 4 or 6 digits, which greatly reduces the user's memory burden and input time, but it also increases security risks and is vulnerable to guessing attacks or shoulder surfing attacks. With the advancement of technology, shoulder surfing attacks have become more difficult to prevent. Surveillance cameras in public places or common smartphones can easily capture the user's unlocking process. Therefore, attackers have changed from shoulder surfing attacks observed by the naked eye to secretly photographing the user's unlocking process and obtaining privacy information from the video. Video-based side channel analysis has been proven to be effective in reconstructing PIN or text-based passwords. Relevant security researchers use trajectory tracking and feature extraction algorithms in the field of computer vision to infer the text entered by the user on the computer keyboard and the PIN lock screen password and pattern password of the mobile phone. The trajectory tracking algorithm can track the user's finger movement trajectory and other targets of interest, analyze the finger movement characteristics, and thus infer the user's input. However, the condition for using the trajectory tracking algorithm is that the video clip needs to be manually cut and the target position needs to be manually marked on the starting frame of the clip. This requires the relevant scheme to manually edit and mark the target to be tracked for each unlocking process in the video. This makes the analysis cost too high, and manual intervention also limits the efficiency of analysis and recognition. It is necessary to find a method that can complete the automatic recognition and target detection of unlocked video clips and automate the steps that require manual participation. This requires the use of a target detection algorithm based on deep learning, and the training of the model is completed by pre-selecting the collected data, so that automatic detection can be achieved. However, the focus of the automatic process is on versatility, which requires the present invention to solve the impact of different mobile phone models and different unlocking postures on recognition, so as to obtain a higher recognition success rate.
[0003] Through the above analysis, the problems and defects of the existing technology are as follows: the existing analysis and recognition methods based on computer vision generally have steps that require human participation, such as video editing, target labeling on the starting frame, and judgment of the device model in the video and query of the keyboard layout. Therefore, the analysis cost is too high, and the recognition efficiency is also limited. Summary of the invention
[0004] In view of the problems existing in the prior art, the present invention provides a method and system for identifying a mobile phone PIN lock screen password based on computer vision.
[0005] The present invention is implemented as follows: a method for identifying a mobile phone PIN lock screen password based on computer vision, the method comprising: recording a video of a user unlocking a mobile phone screen through a PIN code, detecting the positions of fingertips and numeric keys in the video frame, and automatically determining whether it is an unlocking process according to the detection situation; calculating touch points according to the detected fingertip positions and using an optical flow algorithm to assist in tracking the fingertip movement; calculating the overall keyboard layout according to the local numeric key positions, and fine-tuning the keyboard position using the affine mapping principle when there is occlusion; calculating fingertip motion parameters using the optical flow method, and determining the movement or pressing state of the fingertip according to the speed and acceleration; finally, calculating a possible pressed numeric sequence through the position and motion state, and outputting three groups of PIN sequences that the user is most likely to press after the unlocking process is completed.
[0006] Furthermore, the mobile phone PIN lock screen password recognition method based on computer vision specifically includes the following steps:
[0007] Step 1: Obtain a video of the process of a user unlocking the phone screen using a PIN code. The user's finger and unlocking screen must appear clearly in the video. In addition, there are restrictions on the shooting distance, angle, and device model. Therefore, the video can be from a surveillance camera or a common device with a camera function, such as a smartphone.
[0008] Step 2: Detect the fingertips and the positions of the number keys 1, 3, and 8 in the video frame, and automatically determine whether the current frame belongs to the unlocking process based on the detection situation. This step will detect one or more unlocking processes contained in a long video, and complete the recognition and calculation of the input PIN for each process;
[0009] Step 3: For the obtained fingertip position rectangle, calculate the fingertip touch point to achieve more accurate fingertip positioning. Use the optical flow algorithm to assist in tracking the fingertip movement, thereby increasing the detection success rate of the target in the video;
[0010] Step 4: Calculate the overall keyboard layout for the obtained rectangular frames of the number keys 1, 3, and 8. This method uses the spatial layout characteristics of the keyboard by estimating the whole locally, so it is suitable for device recognition of different models or sizes. At the same time, the affine mapping principle is used to fine-tune the keyboard position when there is occlusion.
[0011] Step 5: Use the optical flow method to calculate the fingertip motion parameters, and determine the movement or pressing state of the fingertip based on the speed and acceleration;
[0012] Step 6: When the fingertip touch point is located in the rectangular frame of a key on the numeric keyboard, the probability of the key being pressed is calculated and saved according to the position and fingertip motion parameters. This step uses the calculated position information and motion parameters to judge in two dimensions of time and space to obtain a more accurate key state;
[0013] Step 7: The calculation process is performed frame by frame until the unlocking process is completed. At this time, the three most likely PIN sequences are output based on the saved digital key sequences and their probabilities.
[0014] Furthermore, in step 2, the YOLO series target detection algorithm is used to detect four target positions, namely, fingertips, number keys 1, 3, and 4. The frame in which both the fingertip and number key positions are detected is regarded as the beginning of the unlocking process, and the frame in which the number key positions are not detected for 5 consecutive frames is regarded as the end of the unlocking process.
[0015] Furthermore, in step three, the fingertip position rectangular frame output by the YOLO model is relocated, the image is grayed in turn, the fingertip contour is detected and an ellipse is fitted to the contour, and the center of the ellipse is regarded as the fingertip touch point; and when the accuracy of the model output position is too low, the LK optical flow algorithm is used as an auxiliary tracking method to determine the fingertip position in the time dimension.
[0016] Furthermore, the calculation of step six is a scoring mechanism. When the fingertip touch point passes through a certain numeric key position, the key is recorded and obtains a default score of 0. When the movement of the fingertip shows a trend of first decelerating and then accelerating within the key range, the key is considered to be pressed once and the score of the key is increased by 1. The higher the score, the higher the probability that the key is pressed or pressed multiple times.
[0017] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the mobile phone PIN lock screen password recognition method based on computer vision.
[0018] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to perform the steps of the computer vision-based mobile phone PIN lock screen password recognition method.
[0019] Another object of the present invention is to provide an information data processing terminal, which is used to implement the steps of the mobile phone PIN lock screen password recognition method based on computer vision.
[0020] Another object of the present invention is to provide a mobile phone PIN lock screen password recognition system for implementing the mobile phone PIN lock screen password recognition method based on computer vision, and the mobile phone PIN lock screen password recognition system comprises:
[0021] The target position detection module is used to detect the target position and determine the start and end of the unlocking process. The YOLO series target detection algorithm is used to identify the positions of the four targets, namely the fingertips, the number keys 1, 3, and 8, in each frame of the video. When the fingertips and the number keys are detected at the same time, it is considered that the unlocking process has started. When the number keypad is not detected for 5 consecutive frames, it is considered that the unlocking process has ended.
[0022] The fingertip positioning and tracking module is used to accurately locate and track the fingertip touch points. The edge detection and ellipse fitting algorithm are used to re-locate the fingertip position rectangle output by the YOLO model, and the center of the ellipse is regarded as the fingertip touch point. The LK optical flow algorithm is used as an auxiliary tracking method. When the accuracy of the fingertip rectangle output by YOLO is too low, the optical flow algorithm is used to determine the fingertip position in the time dimension.
[0023] The keyboard layout calculation module is used to calculate the overall layout of the numeric keyboard and estimate the keyboard layout when there is occlusion. It uses the same layout characteristics of the mobile phone numeric keyboard to calculate all the positions of the numeric keys 0 to 9 through the positions of the numeric keys 1, 3, and 8. When there is occlusion, the layout calculated in the previous frame is mapped to the current frame through the affine mapping principle.
[0024] The key dynamic recognition module is used to calculate the finger motion parameters, and judge the finger movement or pressing state according to the motion parameters and position, and finally filter out the PIN sequence entered by the user.
[0025] Another object of the present invention is to provide a mobile device, on which the mobile device is installed with the mobile phone PIN lock screen password recognition system.
[0026] In combination with the above technical solutions and the technical problems solved, please analyze the advantages and positive effects of the technical solutions to be protected by the present invention from the following aspects:
[0027] First, in view of the technical problems existing in the above-mentioned prior art and the difficulty of solving the problems, the technical solutions to be protected by the present invention and the results and data during the research and development process are closely combined to analyze in detail and deeply how the technical solutions of the present invention solve the technical problems, and some creative technical effects brought about after solving the problems. The specific description is as follows:
[0028] The present invention focuses on solving the cost and efficiency problems caused by the need for manual participation in previous video analysis solutions. The introduction of a deep learning-based target detection model greatly reduces labor costs, making the solution capable of large-scale video data analysis and calculation.
[0029] Second, considering the technical solution as a whole or from the perspective of the product, the technical effects and advantages of the technical solution to be protected by the present invention are described in detail as follows:
[0030] The present invention uses a machine learning algorithm to replace the manual processing steps to reduce the recognition cost, and no longer restricts the model of mobile phone devices, thereby improving the versatility of the method, which is of great significance to the research on related security aspects. Under the premise of maintaining the recognition accuracy, the present invention trains the target detection model to complete the automatic detection of fingers and the automatic recognition of unlocking fragments, replacing the manual processing steps to reduce the recognition cost. At the same time, the model of mobile phone devices is no longer restricted, thereby improving the versatility of the method.
[0031] Third, as auxiliary evidence of the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:
[0032] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are:
[0033] The present invention relates to user privacy security and is mainly used for security research or to assist relevant security personnel in completing video analysis.
[0034] (2) The technical solution of the present invention fills the technical gap in the industry at home and abroad:
[0035] Object detection algorithms based on deep learning are widely used in the industry, such as face recognition, unmanned driving, surveillance security, etc. However, the development of technology has two sides. While artificial intelligence technology brings convenience to people's lives, it also brings the risk of privacy leakage.
[0036] The privacy leakage problem caused by artificial intelligence technology has not received widespread attention. In current security research at home and abroad, video-based analysis solutions generally require manual assistance, such as video editing, key point annotation, etc. This invention completes the automation transformation of traditional solutions by training the YOLO model, proving the application potential of artificial intelligence technology in this field. Once the technology is abused, it will pose a huge threat to public privacy.
[0037] On the one hand, this plan reminds the public to be vigilant about the risks of privacy leakage brought about by technological development. On the other hand, it also provides new analysis tools and research ideas for relevant security personnel or security researchers. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1It is a flow chart of a method for identifying a mobile phone PIN lock screen password based on computer vision provided by an embodiment of the present invention;
[0039] Figure 2 It is a structural schematic diagram of a mobile phone PIN lock screen password recognition system based on computer vision provided by an embodiment of the present invention;
[0040] Figure 3 It is a schematic diagram of a mobile phone PIN lock screen password recognition system based on computer vision provided by an embodiment of the present invention;
[0041] Figure 4 This is an experimental scene diagram of a mobile phone PIN lock screen password recognition system based on computer vision provided by an embodiment of the present invention;
[0042] Figure 5 This is an operation effect diagram of a mobile phone PIN lock screen password recognition system based on computer vision provided by an embodiment of the present invention;
[0043] In the figure: 1. Target position detection module; 2. Fingertip positioning and tracking module; 3. Keyboard layout calculation module; 4. Key dynamic recognition module. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0045] 1. Explanatory Examples In order to enable those skilled in the art to fully understand how to implement the present invention, this section provides an illustrative example that expands and describes the technical solution of the claims.
[0046] like Figure 1 As shown, the method for identifying a mobile phone PIN lock screen password based on computer vision provided by an embodiment of the present invention includes the following steps:
[0047] S101: Acquire a video containing a process of a user unlocking a mobile phone screen using a PIN code, where the video may be from a camera or a device with a camera function such as a smart phone;
[0048] S102: Detect the fingertips and the positions of the number keys 1, 3, and 8 appearing in the video frame, and automatically determine whether the current frame belongs to the unlocking process according to the detection situation;
[0049] S103: for the acquired fingertip position rectangular frame, calculate the fingertip touch point and use the optical flow algorithm to assist in tracking the fingertip movement;
[0050] S104: Calculate the overall keyboard layout for the obtained rectangular frames of the positions of the number keys 1, 3, and 8, and fine-tune the keyboard position using the affine mapping principle when there is occlusion;
[0051] S105: Calculate the fingertip motion parameters using the optical flow method, and determine the movement or pressing state of the fingertip according to the speed and acceleration;
[0052] S106: When the fingertip touch point is located in the rectangular frame of a key on the numeric keyboard, the probability of the key being pressed is calculated according to the position and the fingertip motion parameters and saved;
[0053] S107: The calculation process is performed frame by frame until the unlocking process is completed, at which time the three most likely PIN sequences are output based on the saved digital key sequences and their probabilities.
[0054] like Figure 2 As shown, the mobile phone PIN lock screen password recognition system based on computer vision provided by the embodiment of the present invention includes:
[0055] The target position detection module 1 is used to detect the target position and determine the start and end of the unlocking process. The YOLO series target detection algorithm is used to identify the positions of the four targets, namely, fingertips, number keys 1, 3, and 8, in each frame of the video; when the fingertips and number keys are detected at the same time, it is considered the start of the unlocking process, and when the number keypad is not detected for 5 consecutive frames, it is considered the end of the unlocking process.
[0056] The fingertip positioning and tracking module 2 is used to accurately locate and track the fingertip touch points. The fingertip position rectangle output by the YOLO model is relocated using edge detection and ellipse fitting algorithms, and the center of the ellipse is regarded as the fingertip touch point. The LK optical flow algorithm is used as an auxiliary tracking method. When the fingertip rectangle output by YOLO is too inaccurate, the fingertip position is determined in the time dimension using the optical flow algorithm.
[0057] The keyboard layout calculation module 3 is used to calculate the overall layout of the numeric keyboard and estimate the keyboard layout when there is occlusion. Using the same layout characteristics of the mobile phone numeric keyboard, the positions of the numeric keys 1, 3, and 8 are used to calculate all the positions of the numeric keys 0 to 9; when there is occlusion, the layout calculated in the previous frame is mapped to the current frame through the affine mapping principle.
[0058] The key dynamic recognition module 4 is used to calculate the finger motion parameters, and judge the finger movement or pressing state according to the motion parameters and position, and finally filter out the PIN sequence input by the user.
[0059] 2. Application Examples: In order to prove the creativity and technical value of the technical solution of the present invention, this section provides application examples of the technical solution of the claims on specific products or related technologies.
[0060] Surveillance security: This solution can assist security personnel who are authorized to view video surveillance to complete video analysis.
[0061] Privacy protection: For surveillance cameras on the market, when this solution is used to identify the user unlocking process, it can give targeted warnings to users, or perform data desensitization processing such as coding the target location of the source video. 3. Evidence of the effects of the embodiments. The embodiments of the present invention have achieved some positive effects during the research and development or use process, and do have great advantages over the prior art. The following content is described in conjunction with the data, charts, etc. of the test process.
[0062] like Figure 4 As shown, the test video can be from a surveillance camera or a smartphone used for taking pictures, and there may be a clip of the user unlocking the phone in the video. Such a video is used for PIN lock screen password recognition to automatically analyze the PIN sequence entered by the user.
[0063] like Figure 5 As shown in the figure, in the simulation implementation of the scheme, the experimenter uses the index finger to unlock the phone, and the input sequence is 456789. During the unlocking process, the operation status of the relevant modules is output, where the red rectangle is the detected or calculated position range of the numeric key, and the green dot is the fingertip touch point calculated based on the position of the fingertip rectangle. The motion parameters of the fingertip such as speed and acceleration will be calculated during the unlocking process, and the pressed numeric key will be calculated in combination with the position information.
[0064] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.
[0065] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.
Claims
1. A method for identifying a mobile phone PIN lock screen password based on computer vision, characterized in that: The computer vision-based mobile phone PIN lock screen password recognition method includes watching a video of a user unlocking a mobile phone screen using a PIN code, detecting the positions of fingertips and numeric keys appearing in the video frame, and automatically determining whether it is an unlocking process based on the detection situation; calculating the touch points based on the detected fingertip positions and using an optical flow algorithm to assist in tracking the fingertip movement; calculating the overall keyboard layout based on the local numeric key positions, and fine-tuning the keyboard position using the affine mapping principle when there is occlusion; calculating fingertip motion parameters using the optical flow method, and determining the movement or pressing state of the fingertip based on the speed and acceleration; finally, calculating the possible pressed numeric sequence based on the position and motion state, and outputting the three groups of PIN sequences that the user is most likely to press after the unlocking process is completed; The computer vision-based mobile phone PIN lock screen password recognition method specifically includes the following steps: Step 1: Obtain a video of the user unlocking the phone screen using a PIN code. The video can be from a camera or a device with a camera function such as a smartphone. Step 2: Detect the fingertips and the positions of the number keys 1, 3, and 8 in the video frame, and automatically determine whether the current frame belongs to the unlocking process based on the detection situation; Step 3: For the obtained fingertip position rectangle, calculate the fingertip touch point and use the optical flow algorithm to assist in tracking the fingertip movement; Step 4: Calculate the overall keyboard layout for the obtained rectangular frames of the positions of the numeric keys 1, 3, and 8, and use the affine mapping principle to fine-tune the keyboard position when there is occlusion; Step 5: Use the optical flow method to calculate the fingertip motion parameters, and determine the movement or pressing state of the fingertip based on the speed and acceleration; Step 6: When the fingertip touch point is located in the rectangular frame of a key on the numeric keyboard, the probability of the key being pressed is calculated and saved according to the position and fingertip motion parameters; Step 7: The calculation process is performed frame by frame until the unlocking process is completed, at which time the three most likely PIN sequences are output based on the saved digital key sequences and their probabilities; In step 2, the YOLO series target detection algorithm is used to detect four target positions, namely, fingertips, number keys 1, 3, and 4. The frame in which both the fingertips and number key positions are detected is regarded as the beginning of the unlocking process, and the frame in which the number key positions are not detected for five consecutive frames is regarded as the end of the unlocking process.
2. The method for identifying a mobile phone PIN lock screen password based on computer vision as claimed in claim 1, characterized in that: The step three performs secondary positioning on the fingertip position rectangle frame output by the YOLO model, grays the image in turn, detects the fingertip contour and performs ellipse fitting on the contour, and regards the center of the ellipse as the fingertip touch point; and when the accuracy of the model output position is too low, uses the LK optical flow algorithm as an auxiliary tracking method to determine the fingertip position in the time dimension.
3. The method for identifying a mobile phone PIN lock screen password based on computer vision as claimed in claim 1, characterized in that: The calculation of step six is a scoring mechanism. When the fingertip touch point passes through a certain numeric key position, the key is recorded and obtains a default score of 0. When the movement of the fingertip shows a trend of first decelerating and then accelerating within the key range, the key is considered to be pressed once and the score of the key is increased by 1. The higher the score, the higher the probability that the key is pressed or pressed multiple times.
4. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the computer vision-based mobile phone PIN lock screen password recognition method according to any one of claims 1 to 3.
5. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the computer vision-based mobile phone PIN lock screen password recognition method according to any one of claims 1 to 3.
6. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the steps of the computer vision-based mobile phone PIN lock screen password recognition method as described in any one of claims 1 to 3.
7. A mobile phone PIN lock screen password recognition system implementing the mobile phone PIN lock screen password recognition method based on computer vision as described in any one of claims 1 to 5, characterized in that: The mobile phone PIN lock screen password recognition system includes: The target position detection module is used to detect the target position and determine the start and end of the unlocking process. The YOLO series target detection algorithm is used to identify the positions of the four targets, namely the fingertips, the number keys 1, 3, and 8, in each frame of the video. When the fingertips and the number keys are detected at the same time, it is considered that the unlocking process has started. When the number keypad is not detected for 5 consecutive frames, it is considered that the unlocking process has ended. The fingertip positioning and tracking module is used to accurately locate and track the fingertip touch points. The edge detection and ellipse fitting algorithm are used to re-locate the fingertip position rectangle output by the YOLO model, and the center of the ellipse is regarded as the fingertip touch point. The LK optical flow algorithm is used as an auxiliary tracking method. When the accuracy of the fingertip rectangle output by YOLO is too low, the optical flow algorithm is used to determine the fingertip position in the time dimension. The keyboard layout calculation module is used to calculate the overall layout of the numeric keypad and estimate the keyboard layout when there is occlusion. It uses the same layout characteristics of the mobile phone numeric keypad to calculate all the positions of the numeric keys 0 to 9 through the positions of the numeric keys 1, 3, and 8. When there is occlusion, the layout calculated in the previous frame is mapped to the current frame through the affine mapping principle. The key dynamic recognition module is used to calculate the finger motion parameters, and judge the finger movement or pressing state according to the motion parameters and position, and finally filter out the PIN sequence entered by the user.
8. A mobile device, characterized in that: The mobile device is installed with the mobile phone PIN lock screen password recognition system described in claim 7.
Citation Information
Patent Citations
Video-based gesture password track restoration system in complex scene
CN112926518A
Password pattern recognition authentication method
US10540492B1
Systems and Methods of Camera-Based Fingertip Tracking
US20090160767A1
Dynamic Authorization
US20150347725A1
Gesture- and gaze-based visual data acquisition system
US20190246036A1
