首页>
外国专利>
A uniform deep convolution neural network for the estimation of free space, the estimation of the object recognition and the object position
A uniform deep convolution neural network for the estimation of free space, the estimation of the object recognition and the object position
展开▼
机译:用于自由空间估计,对象识别和对象位置估计的统一深度卷积神经网络
展开▼
页面导航
摘要
著录项
相似文献
摘要
A method in a vehicle for performing a plurality of on-vehicle tasks concurrently provided in the same network using deep machine learning algorithms. The method includes obtaining imaging sensor data from a sensor on the vehicle, determining a set of features from the imaging sensor data using a plurality of feature layers in a convolutional neural network, and simultaneously using the convolutional neural network, estimating constraint frames for detected objects, free space boundaries , and object positions for detected objects of the set of features determined by the plurality of feature layers. The neural network may include: a plurality of free space estimation layers configured to determine the free space boundaries in the imaging sensor data, a plurality of object detection layers configured to capture objects in the image, and estimate the bounding frames surrounding the detected objects , and a plurality of object position detection layers configured to estimate the direction of each object.
展开▼