-
公开(公告)号：WO2019200434A1
公开(公告)日：2019-10-24
申请号：PCT/AU2019/050347
申请日：2019-04-18
Applicant: SEEING MACHINES LIMITED
Inventor： EDWARDS, Timothy James Henry
Abstract: Described herein is a monitoring system (100) including one or more infrared light sources (108, 110) for illuminating a subject (102) in a sequenced manner. System (100) also includes a camera (106) for capturing images of the subject (102) during periods in which the subject (102) is illuminated by one of the light sources (108, 110). A processor (118) processes the captured images to determine a brightness measure of the images and a controller (120) controls the output power of the infrared light sources (108, 110) in response to the brightness measure. In response to the processor (118) detecting a brightness measure below a predetermined brightness threshold, the controller (120) is configured to switch off or reduce an output illumination intensity of one of the infrared light sources (108, 110).
-
公开(公告)号：WO2010042990A1
公开(公告)日：2010-04-22
申请号：PCT/AU2009/001360
申请日：2009-10-15
Applicant: SEEING MACHINES LIMITED , EDWARDS, Timothy, James, Henry , LANGDALE-SMITH, Nicholas, John
CPC classification number: G06Q30/02 , G06T19/006
Abstract: Face tracking is used to display a live image of the person being tracked with a synthetically generated 3D model of eyewear (for example, sunglasses or spectacles) overlaid on the person's face to provide a virtual mirror for the person, for example to improve the sales and marketing of eyewear over the internet. The person need not remove their existing glasses; by tracking the outline of the glasses and using 2D image recovery any real glasses are removed and replaced with virtual ones. This ensures the synthetically rendered image is convincing. Optionally, since facial features are tracked in 3D with a high degree of accuracy and precision, the user can also select eyewear based on an appropriate fit to their facial proportions.
Abstract translation: 面部跟踪用于通过合成生成的3D模型（例如，太阳眼镜或眼镜）来覆盖人脸上的身体来显示正在跟踪的人的实况图像，以提供人的虚拟镜像，例如改善销售 并通过互联网营销眼镜。 该人不需要移除现有的眼镜; 通过跟踪眼镜的轮廓和使用2D图像恢复，任何真正的眼镜都被删除，并用虚拟眼镜代替。 这确保了合成渲染的图像是令人信服的。 可选地，由于面部特征以高精度和精确度以3D的方式进行跟踪，所以用户还可以基于适合他们的面部比例来选择眼镜。
-
公开(公告)号：WO2022150874A1
公开(公告)日：2022-07-21
申请号：PCT/AU2021/051563
申请日：2021-12-24
Applicant: SEEING MACHINES LIMITED
IPC: G06V40/40 , G01N21/359 , B60W40/08
Abstract: Described herein is a subject monitoring system (100). The system (100) includes a near infrared illumination source (108) configured to illuminate a scene with infrared light having a spatial beam characteristic to generate a spatial pattern and an image sensor (106) configured to capture one or more images of the scene when illuminated by the illumination source (108). System (100) also includes a processor (112) configured to process the captured one or more images by determining a degree of presence or modification of the spatial pattern by objects in the scene within pixel sub-regions of an image. Processor (112) also classifies one or more pixel sub-regions of the image as including human skin or other material based on the degree of modification of the spatial pattern identified in that pixel sub- region.
-
4.发明申请
公开(公告)号：WO2019084595A1
公开(公告)日：2019-05-09
申请号：PCT/AU2018/050776
申请日：2018-07-27
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John
Abstract: Described herein is a system and method for performing eye tracking. One embodiment provides a system (100) including a camera (106) for capturing images of a vehicle driver's (102) eye and light emitting diodes (LEDs – 108 and 110) configured to selectively illuminate the driver's eye during image capture by the camera (106). A processor (118) is configured to process at least a subset of the captured images to determine one or more eye tracking parameters of the subject's eye and to determine one or more illumination characteristics of the images. A controller (120) is configured to send an LED control signal to the LEDs (108 and 110) to control the drive current amplitude and pulse time of the LEDs (108 and 110). The controller (120) selectively adjusts the drive current amplitude and/or pulse time based on the determined illumination characteristics of a previous captured image or images.
-
公开(公告)号：WO2018000039A1
公开(公告)日：2018-01-04
申请号：PCT/AU2017/050669
申请日：2017-06-29
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John , EDWARDS, Timothy James Henry
Abstract: Described herein is a method (500) of registering the position and orientation of one or more cameras (312-315) in a camera imaging system (100, 400). The method includes: at step (502), receiving, from a depth imaging device (330), data indicative of a three dimensional image of the scene. At step (503) the three dimensional image is calibrated with a reference frame relative to the scene. The reference frame including a reference position and reference orientation. At step (504), a three dimensional position of each of the cameras within the three dimensional image is determined in the reference frame. At step (505), an orientation of each camera is determined in at least one dimension in the reference frame. At step (506) the position and orientation for each camera is combined to determine a camera pose in the reference frame.
Abstract translation: 这里描述的是在相机成像系统（100,400）中记录一个或多个相机（312-315）的位置和方向的方法（500）。 该方法包括：在步骤（502），从深度成像装置（330）接收指示场景的三维图像的数据。 在步骤（503），用相对于场景的参考帧校准三维图像。 参考框架包括参考位置和参考定位。 在步骤（504），在参考帧中确定三维图像内的每个相机的三维位置。 在步骤（505），在参考帧的至少一个维度中确定每个摄像机的方位。 在步骤（506），组合每个相机的位置和方向以确定参考帧中的相机姿态。
-
Patent Agency Ranking6.发明申请SYSTEM AND METHOD FOR IDENTIFYING A CAMERA POSE OF A FORWARD FACING CAMERA IN A VEHICLE 审中-公开用于识别车辆中正面相机的相机位置的系统和方法
公开(公告)号：WO2018000038A1
公开(公告)日：2018-01-04
申请号：PCT/AU2017/050668
申请日：2017-06-29
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John , EDWARDS, Timothy James Henry
IPC: G06T7/80 , G06T7/73 , G06K9/62 , B60R21/015
Abstract: A computerized method of determining a camera pose of a forward facing camera in a vehicle scene, including: capturing images of a vehicle driver's face from a driver facing camera and images of the forward road scene from a forward facing camera; processing the images of the driver's face to derive gaze direction data in a vehicle frame of reference; the gaze direction data are statistically collated into a frequency distribution of gaze angles; peaks in the frequency distribution are identified and associated with reference points in the images of the forward road scene to determine one or more reference gaze positions in the vehicle reference frame; the one or more reference gaze positions are correlated with a position of the reference points in the forward facing camera reference frame to determine a camera pose of a forward facing camera in the vehicle frame of reference.
Abstract translation: 一种确定车辆场景中的前置摄像头的摄像机姿态的计算机化方法，包括：从面向驾驶员的摄像头拍摄车辆驾驶员的脸部图像和从前向摄像头获取前方道路场景的图像 面向相机; 处理驾驶员脸部的图像以导出车辆参考系中的注视方向数据; 将注视方向数据统计地整理为注视角度的频率分布; 识别频率分布中的峰值并将其与前方道路场景的图像中的参考点相关联以确定车辆参考系中的一个或多个参考注视位置; 所述一个或多个参考注视位置与所述前向相机参考帧中的所述参考点的位置相关，以确定所述车辆参照系中的前向相机的相机姿态。
-
公开(公告)号：WO2021087573A1
公开(公告)日：2021-05-14
申请号：PCT/AU2020/051214
申请日：2020-11-06
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John , MCCRACKEN, Fergus
Abstract: Described herein is a method (800) and system for controlling one or more illumination devices in an eye tracker system (100) such that a measured pupil/iris contrast exceeds a predefined minimum pupil/iris contrast. The method (100) includes: a. capturing images of a subject (102), including one or both of the subject's eyes, during predefined image capture periods; b. illuminating, from one or more illumination devices (108 and 110), one or both of the subject's eyes during the predefined image capture periods, wherein at least one illumination device (108 and 110) is located sufficiently close to a lens of the camera to generate bright pupil effects; and c. selectively varying the output power of at least one of the illumination devices (108 and 110) to generate a bright pupil reflection intensity such that a measured pupil/iris contrast in a captured image exceeds a predefined minimum pupil/iris contrast.
-
公开(公告)号：WO2019241834A1
公开(公告)日：2019-12-26
申请号：PCT/AU2019/050620
申请日：2019-06-17
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John
Abstract: Described herein is an image pre-processing system and method. One embodiment provides a method (500) including: at step (501), receiving a plurality of images captured at a first frame rate, the plurality of images captured under at least two different image conditions; pre-processing the plurality of images by: at step (502) identifying one or more regions of interest within the images; at step (503), performing a visibility measure on the one or more regions of interest; and, at step (504), selecting a subset of the plurality of images based on the visibility measure; and, at step (505), feeding the subset of images to an image processing pipeline for subsequent processing at a second frame rate that is lower than the first frame rate.
-
公开(公告)号：WO2019119025A1
公开(公告)日：2019-06-27
申请号：PCT/AU2018/051314
申请日：2018-12-07
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John
Abstract: Described herein is an imaging system (200) for a driver monitoring system (100). The imaging system (200) includes a light source (108) for generating an input light beam (202) and projecting the input light beam (202) along a path towards a driver (102) of a vehicle. System (200) also includes a dielectric metasurface (201) positioned within the path of the input light beam (202). The metasurface (201) has a two dimensional array of surface elements configured to impose predetermined phase, polarization and/or intensity changes to the input light beam (202) to generate an output light beam (204) for illuminating the driver (102). System (200) further includes an image sensor (106) configured to image reflected light (208) being light from the output light beam (204) that is reflected from the driver.
-
公开(公告)号：WO2018000037A1
公开(公告)日：2018-01-04
申请号：PCT/AU2017/050667
申请日：2017-06-29
Applicant: SEEING MACHINES LIMITED
Inventor： NOBLE, John , EDWARDS, Timothy James Henry
Abstract: Described herein are systems and methods of determining a pose of a camera within a vehicle scene. In one embodiment, a method (400) includes the initial step (401) of capturing an image of the vehicle scene from the camera. At step (402), reference data indicative of the vehicle scene is loaded, the reference data includes positions and orientations of known features within the vehicle scene. Next, at step (403), the geometric appearance of one or more of the known features is identified within the image. Finally, at step (404), the three dimensional position and orientation of the camera relative to the known features identified in step c) is determined from the geometric appearance, and a pose of the camera within the vehicle scene is calculated.
Abstract translation: 这里描述的是确定车辆场景内的相机姿态的系统和方法。 在一个实施例中，方法（400）包括从相机捕捉车辆场景的图像的初始步骤（401）。 在步骤（402），加载指示车辆场景的参考数据，参考数据包括车辆场景内的已知特征的位置和取向。 接下来，在步骤（403）中，在图像内识别一个或多个已知特征的几何外观。 最后，在步骤（404），根据几何外观确定摄像机相对于步骤c）中识别的已知特征的三维位置和取向，并且计算车辆场景内摄像机的姿态。
-
-
-
-
-
-
-
-
-
Search Results
25
records
(took 0.012 seconds)
