Annular Panoramic Video Extraction for Flexible Format Generation

The patent divides the 360° annular panoramic video into multiple extraction areas (first extraction area and second extraction area) that can be independently selected and processed. This segmentation allows the system to extract different rectangular regions from the circular panoramic view, enabling multiple video formats (e.g., panoramic mode, normal mode, split-screen mode) to be generated from a single captured frame, thus resolving the contradiction between wide field of view coverage and video format flexibility.

Instead of forcing the user to adapt to the donut-shaped display format, the patent inverts the approach by extracting rectangular regions from the circular panoramic view and displaying them in conventional rectangular formats. This allows users to view specific areas of interest in a natural rectangular format while still having access to the full 360° panoramic content through the extraction area selection mechanism, thereby improving ease of operation without sacrificing panoramic coverage.

The patent performs preliminary action by pre-defining multiple extraction areas (first extraction area, second extraction area, etc.) within the annular panoramic video frame during the capture phase. These extraction areas are predetermined regions that can be directly extracted without requiring complex real-time processing. This preliminary segmentation simplifies subsequent video processing operations, as the system only needs to copy and combine pre-defined regions rather than performing complex calculations, thus achieving video format flexibility while minimizing processing complexity.