Discrete Frame-Based Scene Section Storage Using Timestamps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face high storage costs due to the need to store both discrete frames and scene sections, which is unsustainable given the large volume of data collected during travel.
Innovation Solution
Implementing a discrete frame-based scene section method that uses tags to identify and mark relevant frames, allowing for the storage of start and end frame timestamps to represent scene sections, thereby eliminating the need for dual storage of frames and sections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If both discrete frames and scene sections are stored, then complete data retention is achieved, but storage cost increases significantly
Solution Approach 1:
The patent creates a virtual copy of scene section information by storing only start and end frame timestamps instead of actual frame data. This timestamp-based representation serves as a lightweight proxy that enables scene section identification without duplicating the full frame content, significantly reducing storage requirements while maintaining data integrity
Solution Approach 2:
The patent extracts only the essential temporal boundaries (start and end timestamps) of scene sections from the full frame data. By separating and storing only these critical temporal markers rather than the complete frame sequences, the system achieves efficient storage while preserving the ability to retrieve and reconstruct scene sections when needed
2Measurement precision
If all discrete frames are stored with tags, then accurate scene section identification is achieved, but storage resources are wasted
Solution Approach 1:
Instead of storing tagged copies of actual frame data, the patent creates a virtual representation using timestamps that reference the original frames. This timestamp-based copying mechanism maintains identification accuracy while avoiding the resource overhead of duplicating frame content
Solution Approach 2:
The patent applies tagging selectively only to the boundary frames (start and end frames) of scene sections rather than to all frames within the section. This localized tagging approach maintains precise scene section identification capability while minimizing storage resource consumption by avoiding redundant tags on intermediate frames
3Reliability
If dual storage of frames and sections is implemented, then data completeness is ensured, but storage efficiency decreases
Solution Approach 1:
The patent merges the frame storage system with scene section metadata by using timestamps that directly reference existing frame identifiers. This integration eliminates the need for separate scene section storage while maintaining data completeness, as the timestamp-based references point to the actual frame data already stored in the frame buffer
Solution Approach 2:
The patent creates a lightweight virtual copy of scene section data through timestamp storage rather than duplicating actual frame content. This virtual copying mechanism ensures data completeness can be maintained while dramatically improving storage efficiency by storing only temporal reference information instead of redundant frame data
Data Source
AI summary
A method, apparatus and storage medium for implementing a discrete frame-based scene section are disclosed. The method comprises: obtaining at least one first tag for mapping a discrete frame to a scene section; for any second tag in the at least one first tag, performing polling query on to-be-processed first discrete frames respectively to determine whether there is a second discrete frame matched with the second tag in the first discrete frames; if there is the second discrete frame, marking the second discrete frame with the second tag corresponding to a scene section; if the polling query corresponding to the second tag ends, storing start frame timestamp information and end frame timestamp information of the scene section.