US Patent Application 17731433. SYSTEMS AND METHODS FOR DETECTING TRAFFIC LIGHTS CORRESPONDING TO A DRIVING LANE simplified abstract
SYSTEMS AND METHODS FOR DETECTING TRAFFIC LIGHTS CORRESPONDING TO A DRIVING LANE
Organization Name
TOYOTA JIDOSHA KABUSHIKI KAISHA
Inventor(s)
Kun-Hsin Chen of San Francisco CA (US)
Kuan-Hui Lee of San Jose CA (US)
Chao Fang of Sunnyvale CA (US)
Charles Christopher Ochoa of San Francisco CA (US)
SYSTEMS AND METHODS FOR DETECTING TRAFFIC LIGHTS CORRESPONDING TO A DRIVING LANE - A simplified explanation of the abstract
This abstract first appeared for US patent application 17731433 titled 'SYSTEMS AND METHODS FOR DETECTING TRAFFIC LIGHTS CORRESPONDING TO A DRIVING LANE
Simplified Explanation
This patent application describes a system and method for detecting traffic lights using multiple cameras. Here are the key points:
- The system uses images from multiple cameras to estimate the positions and confidence levels of traffic lights in a vehicle's driving lane.
- A second model aggregates the confidence levels and creates a multi-view stereo composition based on the geometric representations of the traffic lights' positions.
- The second model assigns a relevancy score to each candidate traffic light based on the aggregation, determining how relevant it is to the driving lane.
- The vehicle then executes a task based on the relevancy score, potentially adjusting its driving behavior according to the detected traffic lights.
Original Abstract Submitted
System, methods, and other embodiments described herein relate to detection of traffic lights corresponding to a driving lane from views captured by multiple cameras. In one embodiment, a method includes estimating, by a first model using images from multiple cameras, positions and state confidences of traffic lights corresponding to a driving lane of a vehicle. The method also includes aggregating, by a second model, the state confidences and a multi-view stereo composition from geometric representations associated with the positions of the traffic lights. The method also includes assigning, by the second model according to the aggregating, a relevancy score computed for a candidate traffic light of the traffic lights to the driving lane. The method also includes executing a task by the vehicle according to the relevancy score.
