Panoramic Video Editing via Salient Object Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Watching panoramic videos on flat panel displays requires constant rotation to view salient objects, making the experience cumbersome due to the inability to display images at the correct viewing angle, which affects user engagement.

Innovation Solution

A method for editing panoramic videos by recording the viewing angle of a panoramic camera, performing frame extraction, salient object detection, and tracking using a preset algorithm to generate an object video with adjusted playback speed based on the presence and duration of salient objects, ensuring seamless and engaging content playback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the display constantly rotates to show salient objects in panoramic video, then the viewing angle coverage is improved, but the operation complexity increases and user experience deteriorates

Engineering Contradiction:
Improveviewing angle coverageVSAvoidoperation complexity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system performs preliminary detection of salient objects and pre-calculates the optimal viewing angles during video processing. This allows the playback device to directly display pre-positioned frames showing salient objects without requiring real-time rotation operations by the user, thus resolving the contradiction between comprehensive viewing angle coverage and ease of operation.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If the display shows the forward direction of the panoramic camera, then the operational simplicity is maintained, but the salient objects may be invisible at certain viewing angles

Engineering Contradiction:
Improveoperational simplicityVSAvoidsalient object visibility
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system incorporates feedback mechanisms where salient object detection results from video analysis inform the playback behavior. The detected salient objects and their optimal viewing angles are fed back into the playback system, which then automatically adjusts the displayed viewing angles to ensure salient objects remain visible, thus resolving the contradiction between operational simplicity and salient object visibility.

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If the panoramic video is edited to focus on salient objects with variable playback speed, then the user engagement is improved, but the processing complexity increases

Engineering Contradiction:
Improveuser engagementVSAvoidprocessing complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system implements self-service through automated salient object detection and tracking algorithms that independently analyze video content and determine optimal playback parameters. The system serves itself by automatically identifying salient objects, calculating their viewing angles, and adjusting playback speed without requiring manual intervention, thus resolving the contradiction between enhanced user engagement and reduced processing complexity.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS12143667B2Panorama video editing method, apparatus,device and storage medium
Publication Date: 2024.11.12 ARASHI VISION INC
  • US12143667B2 patent drawing
  • US12143667B2 patent drawing

AI summary

The present invention is applicable to the technical field of panoramic videos. Provided are a panoramic video clip method, apparatus and device, and a storage medium. The method comprises acquiring a panoramic video photographed by means of a panoramic camera, and recording an advancing-direction viewing angle of the panoramic camera during moving photographing; carrying out a frame extraction operation on the acquired panoramic video to obtain a corresponding panoramic video frame, carrying out significant target detection on the panoramic video frame, tracking a detected significant target by using a preset target tracking algorithm, and acquiring a viewing angle at which the tracked significant target is located; and clipping the panoramic video according to the advancing-direction viewing angle and the viewing angle at which the significant target is located, so as to generate a target video corresponding to the panoramic video. A panoramic video is automatically clipped, and the fluency of transitions, and the effectiveness and the degree to which content in a target video is interesting are also ensured.