Video Object Tracking via MSER Part Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video tracking systems face errors in grouping and separating moving objects due to overlapping paths and temporary changes, leading to incorrect classification and labeling of objects in video sequences.
Innovation Solution
A method and system that utilize Maximally Stable Extremal Regions (MSER) detection, part tracking, and clustering to identify moving objects, employing probability functions and hysteresis filtering to stabilize object labeling and correct erroneous groupings, ensuring robustness to scale and frame rate variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If feature detection and path matching are used to track objects in video, then object tracking capability is achieved, but classification errors occur due to overlapping paths
Solution Approach 1:
The patent segments the tracking problem into multiple independent components: feature detection, path extraction, clustering analysis, and classification. By dividing the complex tracking task into these manageable segments, the system can apply specialized algorithms to each component, improving overall classification accuracy while maintaining manageable system complexity.
Solution Approach 2:
The patent introduces clustering analysis as an intermediary step between path matching and final classification. This intermediary process analyzes the spatial and temporal relationships among multiple paths to determine whether they belong to the same object or different objects, thereby resolving classification errors caused by overlapping paths without requiring direct modification of the basic tracking mechanism.
2Measurement precision
If path grouping is performed to identify moving objects, then object identification is achieved, but erroneous separation occurs into multiple objects
Solution Approach 1:
The patent implements a feedback mechanism where clustering results are used to refine and correct path groupings. The clustering analysis continuously evaluates the consistency of path groupings across multiple frames and provides feedback to adjust groupings that appear erroneous, thereby improving identification accuracy while maintaining grouping stability.
Solution Approach 2:
The patent performs preliminary clustering analysis on paths before final object identification is made. By analyzing path relationships in advance and establishing probable groupings before final classification, the system can prevent erroneous separations and ensure more accurate object identification while maintaining stable groupings throughout the video sequence.
3Duration of action of moving object
If feature matching is performed across frames to track objects, then continuous tracking is achieved, but errors occur due to temporary changes in object appearance
Solution Approach 1:
The patent employs dynamic clustering thresholds and adaptive matching criteria that adjust based on the temporal context and object behavior patterns. This dynamic approach allows the system to maintain continuous tracking over extended durations while adapting to temporary changes in object appearance, such as occlusions or pose variations, without sacrificing identification accuracy.
Data Source
AI summary
In one embodiment, a method of determine and track moving objects in a video, including detecting and extracting regions from accepted frames of a video, matching parts including using the extracted parts of a current frame and matching each part from a previous frame to a region in the current frame, tracking the matched parts to form part tracks, and determining a set of path features for each tracked part path. The determined path features are used to classify each path as that of mover or a static. The method includes clustering the paths of movers, including grouping parts of movers that likely belong to a single object, in order to generate one or more single moving objects and track moving objects. Also a system to carry out the method and a non-transitory computer-readable medium that when executed in a processing system causes carrying out the method.