Qualcomm Incorporated

Kneron (taiwan) Co., Ltd.

Hung-Hsin Wu

Karthikeyan Shanmugavadivelu

Adarsh Abhay Golikeri

Shizhong Liu

Wan Shun Vincent Ma

Venkata Ravi Kiran Dayana

Ying Chen Lou

Narayana Karthik Sadanandam Ravirala

William T Frantz

Jose Roberto Menendez

Chiu-Ling Chen

Bike Xie

John D Eaton

Chuqiao Song

A Fast Non-Maximum Suppression (NMS) Algorithm post-processing for object detection includes getting original data output from a deep learning model inference output, the original data including a plurality of bounding boxes, pre-emptively filtering out at least one bounding box of the plurality of bounding boxes from further consideration when applying the algorithm, the at least one bounding box filtered out according to a predetermined criteria, processing data, using sigmoid functions or exponential functions, from bounding boxes of the plurality of bounding boxes not filtered out to generate processed bounding boxes, calculating final scores of the processed bounding boxes, and choosing a processed bounding boxes utilizing the final scores.

Fast non-maximum suppression algorithm for object detection

Method and systems for autofocus triggering focusing are disclosed herein. In one example, a system may include a lens, a memory component configured to store lens parameters of the lens and regions of focus corresponding to the lens parameters, and a processor coupled to the memory and the lens. The processor may be configured to focus the lens on a target object at a first instance of time, receive information indicative of distances from an imaging device to the target object over a period of time, obtain lens parameters of the lens, and determine a region of focus, and trigger the lens to re-focus on the target object if the distance to the target object indicates the target object is outside of the region of focus and the distance to the target object is unchanged for a designated time period.

Systems and methods for autofocus trigger

Described is a method and apparatus for unattended image capture that can identify subjects or faces within an image captured with an image sensor. The methods and apparatus may then score the image based, at least in part, on scores of detected subjects or faces in the image, scores of facial expressions, a focus score, exposure score, stability score, or audio score. If the score of the image is above a threshold, a snapshot image may be stored to a data store on the imaging device. In some aspects, if the score of the image is below a threshold, one or more audible prompts may be generated indicating that subjects should change positions, smile or remain more still during the image capture process.

Method and apparatus for unattended image capture

Described is a method and apparatus for unattended image capture that can identify subjects or faces within an image captured with an image sensor. The methods and apparatus may then score the image based, at least in part, on scores of detected subjects or faces in the image, scores of facial expressions, a focus score, exposure score, stability score, or audio score. If the score of the image is above a threshold, a snapshot image may be stored to a data store on the imaging device. If the score of the image is below a threshold, one or more audible prompts may be generated indicating that subjects should change positions, smile or remain more still during the image capture process.

Described herein are methods and devices that employ a dual shutter button feature associated with an image capture device to recommend a capture mode to a user based on one or more parameters analyzed by the image capture system. As described, providing a primary shutter button and a secondary shutter button enables the user to capture in both a standard capture mode by using the primary shutter button and in an alternate mode by using a secondary shutter button.

Method and apparatus for dual camera shutter

A mobile device uses sensor data related to the type of surface in contact with the mobile device to determine an action to perform. The sensors, by way of example, may be one or more of a microphone and noise generator, a light based proximity sensor, and pressure sensors, such as dielectric elastomers, configured to detect a texture of the surface, and/or pressure waves produced by setting the mobile device down or by a noise generator and reflected by the surface. The mobile device may identify the type of surface and perform the action based on the type of surface. The mobile device may further determine its location based on the sensor data and use that location to identify the action to be performed. The location may be determined using additional data, e.g., data not related to determining the type of surface with which the mobile device is in contact.

Growing community of inventors

San Diego, CA, United States of America

Hung-Hsin Wu