Qualcomm Incorporated

National University of Singapore

Xiaopeng Zhang

Liang Shen

Gang Qiu

Xiaoming Zhou

Edwin Chongwoo Park

Evgeni Petrovich Gousev

Victor Hokkiu Chan

Chen Feng

Jilai Wang

Shaojie Zhuo

Milivoje M Aleksic

Sundeep Vaddadi

Jiaya Jia

Wentao Xu

Xiaokang Zhang

Farrukh Qurashi

Terence Mong Cheng Sim

Sensing of scene-based occurrences is disclosed. In one example, a vision sensor system comprises (1) dedicated computer vision (CV) computation hardware configured to receive sensor data from at least one sensor array and capable of computing CV features using readings from multiple neighboring sensor pixels and (2) a first processing unit communicatively coupled with the dedicated CV computation hardware. The vision sensor system is configured to, in response to processing of the one or more computed CV features indicating a presence of one or more irises in a scene captured by the at least one sensor array, generate data in support of iris-related operations to be performed by a second processing unit and send the generated data to the second processing unit.

Low power data generation for iris-related detection and authentication

Use of pupillary response to visible light for iris authentication is disclosed. One example involves (a) capturing an initial image of an eye including an inner circular boundary between a pupil region and an iris region and an outer circular boundary between the iris region and a sclera region, (b) determining a first size measurement indicative of the inner circular boundary, (c) responsive to at least the first size measurement, modulating one or more visible light sources to output visible light toward the eye of the user, (d) capturing a subsequent image of the eye of the user during a period of pupillary response, (e) obtaining an iris data record of the user, and (f) comparing the iris data record of the user to one or more registered iris data records to authenticate the user.

Light source modulation for iris size adjustment

Systems and methods for detecting a live human face in an image are disclosed. The methods and systems are capable of receiving multispectral image data that includes a human face. The multispectral image data can comprise visible light image data and near-infrared (NIR) image data. The multispectral image data can be processed to detect the human face. The detected human face in the visible light image data can be associated with the detected human face in the NIR image data to determine whether the detected human face is a live human face.

Systems and method for facial verification

Systems and methods for multispectral imaging are disclosed. The multispectral imaging system can include a near infrared (NIR) imaging sensor and a visible imaging sensor. The disclosed systems and methods can be implemented to improve alignment between the NIR and visible images. Once the NIR and visible images are aligned, various types of multispectral processing techniques can be performed on the aligned images.

Multispectral imaging system

Described herein is a system and method for adjusting the orientation of captured video utilizing the data received from a built-in inertial measurement unit such as an accelerometer. During video capture, the device may be held in a position that is not fully vertical or horizontal, and thus not match the true orientation of a scene. This can cause the captured video to appear rotated during playback at the same angle of rotation as the video capture device. The described system can adjust the playback orientation of the rotated video by utilizing sensor data captured during the same time as the video was captured.

System and method for adjusting orientation of captured video

Systems and methods for multispectral imaging are disclosed. The multispectral imaging system can include a near infrared (NIR) imaging sensor and a visible imaging sensor. The disclosed systems and methods can be implemented to de-noise a visible light image using a gradient scale map generated from gradient vectors in the visible light image and a NIR image. The gradient scale map may be used to determine the amount of de-noising guidance applied from the NIR image to the visible light image on a pixel-by-pixel basis.

Near infrared guided image denoising

Near infra-red images of natural scenes usually have better contrast and contain rich texture details that may not be perceived in visible light photographs. The contrast and rich texture details form a NIR image corresponding to a visible light image are useful for enhancing the visual quality of the visible light image. To enhance the visual quality of a visible light image using its corresponding near infra-red image, a computer-implemented method computes a weight region mask from the visible light image, transfers contrast data and texture data from the near infra-red image to the visible light image guided by the weighted region mask. The contrast data is computed from the low frequency subbands of the visible light image and corresponding infra-red image after a wavelet transform by matching the histogram of gradient magnitude. The texture data is computed from the high frequency subbands of both images after wavelet transform.

Growing community of inventors

Richmond Hill, Canada

Xiaopeng Zhang