Joint Object Tracking and Shape Estimation for Unknown Classes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in reliably detecting and tracking objects that do not belong to known classes, such as unusual objects encountered by autonomous vehicles, which can hinder effective collision avoidance and navigation.
Innovation Solution
The proposed solution involves an apparatus and method for joint tracking and shape estimation of objects. This is achieved by detecting features in multiple frames, determining sets of 3D points based on these features and distance information, combining these points, and estimating the shape of the object based on the combined set of 3D points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional object detection methods are used to identify objects in known classes, then detection accuracy for common objects is improved, but the system fails to detect and track unknown or unusual objects that do not fit predefined categories
Solution Approach 1:
The system performs multiple functions: it detects features in images, tracks objects across frames, estimates 3D shapes, and classifies objects. By making the detection system universal rather than specialized for specific object classes, it can handle both known and unknown objects effectively
Solution Approach 2:
The system changes from using fixed object class categories to using continuous 3D shape parameters and feature descriptors. This parameter transformation allows the system to represent and compare objects based on their geometric properties rather than predefined labels, enabling detection of novel objects
2Reliability
If the system attempts to detect and identify all objects in the environment, then collision avoidance capability is improved, but computational complexity and processing time increase
Solution Approach 1:
The system extracts only the essential features needed for tracking and shape estimation from the full image data, rather than processing all visual information. This selective extraction reduces computational complexity while maintaining the ability to detect and track relevant objects for collision avoidance
Solution Approach 2:
The processing system is segmented into distinct modules: feature detection, object tracking, 3D shape estimation, and classification. This modular segmentation allows each component to operate independently and efficiently, reducing overall system complexity while maintaining comprehensive object detection capabilities
Data Source
AI summary
Techniques and systems are provided for shape estimation. For instance, a process can include: detecting first features of an object in a first frame of an environment, the environment including the object; determining a first set of three-dimensional (3D) points for the first frame based on the detected first features and first distance information obtained for the object; detecting second features of the object in a second frame of the environment; determining a second set of 3D points for the second frame based on the detected second features and second distance information obtained for the object; combining the first set of 3D points and the second set of 3D points to generate a combined set of 3D points; and estimating a shape of the object based on the combined set of 3D points.