Visual Object Tracking on Low-End UAVs Using Correlation Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing consumer UAVs face challenges in implementing high-performance visual tracking without GPS, as current solutions require powerful computational platforms like Intel Movidius Myriad VPU, making them unsuitable for low-end consumer UAVs.
Innovation Solution
A real-time visual object tracking system that uses correlation maps computed between image patches within search windows to determine the updated location of a target object, employing techniques such as Fast Fourier Transform and Gaussian kernel auto-correlation, and features like geometry-based and color-based feature maps, suitable for low-end platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high-complexity visual tracking functionalities are implemented, then tracking performance is improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent segments the visual tracking process into distinct modules: feature extraction, correlation map computation, and target location determination. Each module performs a specific function with optimized computational requirements, allowing the system to achieve high tracking performance without requiring a single powerful computational platform like Intel Movidius Myriad VPU.
Solution Approach 2:
The patent changes the computational parameters by using correlation maps computed from image patches within search windows instead of full-frame processing. This parameter change significantly reduces computational complexity while maintaining tracking accuracy, making the system suitable for low-end consumer UAV platforms.
2Measurement precision
If GPS-based tracking is used, then tracking accuracy is improved, but adaptability to GPS-denied environments deteriorates
Solution Approach 1:
The patent creates a visual copy of the target object through correlation maps that represent the spatial relationship between image patches. This visual copy allows the system to track the target using only visual information from the camera, eliminating dependence on GPS signals while maintaining tracking capability in GPS-denied environments.
3Adaptability or versatility
If real-time visual tracking is implemented without GPS, then adaptability to various environments is improved, but computational load increases
Solution Approach 1:
The patent applies local quality by focusing computational resources only on specific regions of interest within the video frames. By defining search windows around the target object and computing correlation maps only for these localized regions rather than processing entire frames, the system achieves environmental adaptability while significantly reducing computational energy consumption suitable for low-end UAV platforms.
Data Source
AI summary
Embodiments described herein provide examples of a real-time visual object tracking system. In one aspect, an unmanned aerial vehicle (UAV) capable of performing real-time visual tracking of a moving object includes: a processor; a memory coupled to the processor; and a camera to capture a video of the moving object. This UAV additionally includes a visual tracking module to: receive a first video image and a first location of the object; receive a second video image following the first video image; place a first search window in the first video image and a second search window in the second video image centered on a second location in the second video image having the same coordinates as the first location; compute a correlation between an image patch within the first search window and an image patch within the second search window; and determine an updated location of the object in the second video image.