Large Format Video Archival System Using Overlapping Tile Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital video systems face challenges in efficiently archiving, storing, and retrieving large format videos due to their high pixel density, which requires significant resources and leads to performance degradation when displaying partial frames, especially when zooming or panning.
Innovation Solution
A system that formats large video frames into overlapping tiles, allowing for efficient storage and retrieval by storing each tile as a contiguous stream, enabling seamless zooming and panning without image degradation, and optimizing storage by using motion compensation and variable length records.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If large format video frames are stored as single continuous data, then storage simplicity is maintained, but retrieval and processing performance degrades due to high pixel density and large size
Solution Approach 1:
The patent divides large format video frames into multiple smaller tiles (e.g., 4k or 8k resolution tiles) that can be independently stored, retrieved, and processed. This segmentation allows the system to handle massive datasets by breaking them into manageable units, improving retrieval performance while maintaining acceptable structural complexity through standardized tiling protocols.
2Speed
If video frames are displayed at full resolution, then image quality is maintained, but processing speed and system response time degrade due to high pixel density
Solution Approach 1:
The patent applies different resolution levels to different regions of the video frame based on display requirements. When users need detailed views, only the relevant tiles are retrieved and displayed at full resolution. For overview or lower-detail requirements, tiles can be displayed at reduced resolution, optimizing processing speed while maintaining image quality where needed.
Solution Approach 2:
The system dynamically adjusts the resolution and retrieval of tiles based on real-time user interactions and display requirements. When zooming or panning occurs, the system dynamically loads appropriate tiles at suitable resolutions, balancing processing speed and image quality adaptively rather than statically.
3Measurement precision
If overlapping tiles are used for seamless zooming and panning, then image quality is maintained during manipulation, but storage space requirements increase
Solution Approach 1:
The patent implements overlapping tiles with controlled overlap dimensions (e.g., 10-20% overlap) that provide just enough redundancy for seamless transitions during zooming and panning. This partial excessive action ensures image quality maintenance during manipulation while minimizing unnecessary storage consumption by avoiding excessive overlap.
4Productivity
If motion compensation is applied to optimize storage, then storage efficiency improves, but processing complexity increases
Solution Approach 1:
The patent applies motion compensation and predictive coding during the initial encoding and storage phase, pre-computing motion vectors and predicting future frame data. This preliminary action optimizes storage efficiency by reducing redundant data during capture, while the complexity is managed through automated processing that occurs during encoding rather than during playback or retrieval.
Data Source
AI summary
A method and system for storing a video on a storage device are provided. The method includes formatting each image in a plurality of images into a plurality of tiles, the plurality of images being captured as a temporal sequence of images at successive points in time. The method further includes selecting a tile from each image in the temporal sequence of images to obtain a temporal sequence of tiles to generate a video segment; selecting another tile from each image in the temporal sequence of images to obtain another temporal sequence of tiles to generate another video segment; and repeating the selecting a tile from each image in the temporal sequence of images to obtain a plurality of temporal sequences of tiles to generate a plurality of video segments. The obtained plurality of video segments are stored in a file on the storage device.