Multi-Store Object Tracker Using Segmented Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object tracking devices face challenges in handling complex tracking scenarios due to variations in object appearance, occlusion, and fast appearance changes, with short-term trackers being inefficient in complex environments and long-term trackers being conservative in updating appearance models.
Innovation Solution
The multi-store tracker employs a dual-component approach based on the Atkinson-Shiffrin memory model, combining a short-term memory store using integrated correlation filters for spatiotemporal consistency and a long-term memory store with key-point matching and RANSAC estimation, which updates key-point features based on a forgetting curve model to adapt to changing object appearances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If short-term tracking devices are used, then tracking speed and efficiency are improved, but tracking reliability deteriorates in complex environments with occlusion and appearance changes
Solution Approach 1:
The tracking system is segmented into two independent components: a short-term tracker for fast tracking and a long-term tracker for reliable appearance modeling. Each component operates with its own memory model and update mechanism, allowing them to function independently while contributing to overall system performance. The short-term tracker handles rapid tracking with incremental updates, while the long-term tracker ensures reliability through conservative appearance model updates.
Solution Approach 2:
The system merges the outputs of the short-term tracker and long-term tracker through a consistency check mechanism. When the short-term tracking result is consistent with the long-term appearance model, the short-term result is accepted for high-speed tracking. When inconsistency is detected, the system switches to long-term tracking results, ensuring reliability while maintaining overall system efficiency.
2Reliability
If long-term tracking devices are used, then tracking reliability is improved, but tracking speed deteriorates due to conservative appearance model updates
Solution Approach 1:
The system dynamically switches between short-term and long-term tracking modes based on consistency checks. The long-term tracker maintains a conservative appearance model that is updated only when necessary, while the short-term tracker provides fast tracking updates. This dynamic switching mechanism allows the system to achieve both reliability and speed by selecting the appropriate tracking mode for each situation.
Solution Approach 2:
The consistency check mechanism acts as an intermediary between the short-term and long-term trackers. It evaluates whether the short-term tracking results are consistent with the long-term appearance model and mediates the selection of final tracking results. This intermediary mechanism enables the system to leverage the speed of short-term tracking while ensuring reliability through long-term appearance verification.
3Adaptability or versatility
If appearance model updates are performed frequently, then adaptability to appearance changes is improved, but measurement precision deteriorates due to noise and false positives
Solution Approach 1:
The long-term tracker performs appearance model updates periodically rather than continuously, only when consistency checks indicate that updates are necessary. This periodic update mechanism prevents overfitting to noisy data while maintaining adaptability to genuine appearance changes. The system balances adaptability and precision by updating the appearance model at appropriate intervals based on tracking consistency.
Data Source
AI summary
An object tracking device includes a short-term processing portion and a long short-term processing portion that are implemented by circuitry and work in a collaborative manner to track an object. The short-term processing portion includes a filter that tracks the object based on short-term memory and spatiotemporal consistency. The long short-term processing portion performs key-point matching-tracking and estimation based on a key-point database in order to track the object. A controller determines an output of the object tracking device based on the processing conducted by the short-term and long short-term processing portions of the tracking device, respectively.