Trajectory-Based Image Search for Moving Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image search technologies fail to identify moving objects in videos based on user-input features, such as trajectory shapes, even when the object is not directly captured in the video.
Innovation Solution
An image search system that includes a video taking unit, a search unit with a storage unit to hold trajectory information, and a display unit, allowing users to input trajectory shapes for searching similar trajectories and displaying corresponding moving object images, regardless of whether the target object is captured in the video.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image search is performed based on captured video frames only, then search accuracy for directly captured objects is improved, but search capability for objects not captured in video is lost
Solution Approach 1:
The patent transitions from searching based solely on spatial image data to incorporating temporal trajectory information as an additional dimension. By extracting movement trajectories (temporal dimension) from video frames and using them as search keys, the system can identify objects even when not directly captured, thus expanding search capability while maintaining accuracy through multi-dimensional matching
Solution Approach 2:
The patent introduces trajectory information as an intermediary between the user's search intent and the actual object identification. When an object is not directly captured, the trajectory serves as a mediator that bridges the gap between available video data and the target object, enabling indirect identification through movement pattern matching
2Adaptability or versatility
If trajectory information is extracted and stored for all moving objects, then search capability for moving objects is improved, but data storage requirements and processing complexity increase
Solution Approach 1:
The patent extracts only the essential trajectory information (movement paths and patterns) from video data, separating this key feature from the complete video content. This extraction approach enables search capability based on movement patterns while reducing the complexity of processing and storing entire video sequences
Solution Approach 2:
The system performs preliminary extraction and storage of trajectory information from video data before the actual search operation. By pre-processing the video to extract movement patterns and storing them in an accessible format, the system reduces the complexity of real-time search operations while maintaining comprehensive search capability
Data Source
AI summary
An image search system comprising an video taking unit that takes a video, a search unit that contains a processor and a storage unit and searches for an image contained in the video, an input unit that receives information that is input to the search unit, and a display unit that displays information that is output from the search unit. The storage unit holds information indicating the shapes of trajectories that represent the travel paths of a plurality of moving objects extracted from the video. When information indicating the shape of a trajectory is input from the input unit, the search unit searches the plurality of moving-object trajectories held in the storage unit for trajectories that are similar to the input trajectory shape, and if such a trajectory is found, the search unit displays, on the display unit, images of the moving objects that have traveled along the found trajectories.