Difference between revisions of "LANE MARKER RECOGNITION: abstract simplified (17655500)"

From WikiPatents
Jump to navigation Jump to search
(Creating a new page)
 
(Creating a new page)
Line 1: Line 1:
The abstract describes a technique for detecting lane markers using a convolutional neural network. The process involves generating feature tensors from an input image, which are then used to generate localizations, horizontal positions, and end positions. Based on these generated positions, the technique determines the positions of the lane markers.
+
The abstract describes a technique for detecting lane markers using a convolutional neural network. It involves generating feature tensors from an input image and processing them through different networks to generate sets of localizations, horizontal positions, and end positions. These sets are then used to determine the positions of lane markers.

Revision as of 16:12, 1 October 2023

The abstract describes a technique for detecting lane markers using a convolutional neural network. It involves generating feature tensors from an input image and processing them through different networks to generate sets of localizations, horizontal positions, and end positions. These sets are then used to determine the positions of lane markers.