Qualcomm incorporated (20240202949). DEPTH ESTIMATION FOR MONOCULAR SYSTEMS USING ADAPTIVE GROUND TRUTH WEIGHTING simplified abstract
Contents
DEPTH ESTIMATION FOR MONOCULAR SYSTEMS USING ADAPTIVE GROUND TRUTH WEIGHTING
Organization Name
Inventor(s)
Amin Ansari of Federal Way WA (US)
Sai Madhuraj Jadhav of San Diego CA (US)
Yunxiao Shi of San Diego CA (US)
Gautam Sachdeva of San Diego CA (US)
Avdhut Joshi of San Marcos CA (US)
DEPTH ESTIMATION FOR MONOCULAR SYSTEMS USING ADAPTIVE GROUND TRUTH WEIGHTING - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240202949 titled 'DEPTH ESTIMATION FOR MONOCULAR SYSTEMS USING ADAPTIVE GROUND TRUTH WEIGHTING
Simplified Explanation:
This patent application discusses systems, methods, and devices for vehicle driving assistance systems that utilize image processing. The computing device receives predicted and measured depth maps, calculates the time difference between them, and determines a supervision loss term based on this difference to train a model for generating predicted depth maps.
- The patent application focuses on vehicle driving assistance systems that incorporate image processing.
- The computing device receives predicted and measured depth maps and calculates the time difference between them.
- A supervision loss term is determined based on the time difference to train a model for generating predicted depth maps.
Potential Applications: This technology can be applied in autonomous vehicles, advanced driver assistance systems (ADAS), and other automotive safety systems that rely on image processing for depth perception.
Problems Solved: This technology addresses the need for accurate depth perception in vehicle driving assistance systems, improving the safety and efficiency of autonomous and semi-autonomous vehicles.
Benefits: - Enhanced accuracy in depth perception for vehicle driving assistance systems - Improved safety and efficiency in autonomous and semi-autonomous vehicles - Better performance of advanced driver assistance systems (ADAS)
Commercial Applications: Title: Advanced Image Processing for Vehicle Driving Assistance Systems This technology can be utilized by automotive manufacturers, technology companies, and research institutions to develop advanced driver assistance systems for commercial vehicles, passenger cars, and other automotive applications. The market implications include improved safety features, increased automation in vehicles, and enhanced driving experiences for consumers.
Questions about Image Processing for Vehicle Driving Assistance Systems: 1. How does the time difference between predicted and measured depth maps impact the training of the model? 2. What are the potential challenges in implementing this technology in real-world driving scenarios?
Original Abstract Submitted
this disclosure provides systems, methods, and devices for vehicle driving assistance systems that support image processing. in a first aspect, a computing device may receive a predicted depth map and a measured depth map and may determine a time difference between the predicted depth map and the measured depth map. a supervision loss term may be determined based on the time difference, such as by weighting the supervision loss term based on the time difference. the computing device may train a model based on the supervision loss term, such as a model that generated the predicted depth map. other aspects and features are also claimed and described.
