Variable Rate Time-Lapse Video Saliency for Security Footage Review
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Home video monitoring systems generate large amounts of unnecessary footage due to false motion detection from weather events and outdoor lighting, requiring users to sift through uninteresting clips, which is inefficient and time-consuming.
Innovation Solution
A method for generating a time-lapse video representation with variable speeds, where motion events of higher priority are played back at a slower frame rate than non-motion events, allowing users to quickly identify important moments while skipping uninteresting footage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If motion-activated recording is used to reduce unnecessary footage, then the quantity of recorded footage is reduced, but false motion detection from weather events and outdoor lighting generates uninteresting clips that still require review
Solution Approach 1:
The patent segments the time-lapse video into multiple portions with different frame rates based on detected motion events. High-priority motion events (e.g., human activity) are displayed at slower frame rates for detailed review, while low-priority events (e.g., weather-related motion) are displayed at faster frame rates for quick scanning. This segmentation allows users to efficiently review uninteresting footage while maintaining the ability to examine important moments in detail.
2Reliability
If continuous video recording is used to capture all events, then no footage is missed, but users must sift through voluminous quantities of footage to identify important moments
Solution Approach 1:
The patent dynamically adjusts the frame rate of the time-lapse video based on the priority of detected motion events. The system continuously monitors video footage, detects motion events, determines their priority levels, and adjusts playback speed accordingly. This dynamic approach maintains reliability by capturing all events while reducing review time through adaptive speed variation.
Solution Approach 2:
The patent changes the temporal parameter (frame rate) of the video based on the priority of motion events. High-priority events are played back at slower frame rates (e.g., 1fps or 0.5fps) to allow detailed examination, while low-priority events are played back at faster frame rates (e.g., 4fps or 10fps) to enable quick scanning. This parameter change resolves the contradiction between complete event capture and efficient review.
3Productivity
If variable frame rate time-lapse is used to highlight important events, then review efficiency is improved, but the system complexity increases due to motion event detection and priority determination
Solution Approach 1:
The patent implements a self-service system where the video processing system automatically detects motion events, determines their priority levels, and adjusts frame rates without requiring user intervention. The system uses built-in motion detection algorithms and pre-defined priority criteria (e.g., human detection, animal detection, weather events) to autonomously create the variable frame rate time-lapse, reducing the need for complex user configuration.
4Volume of stationary object
If motion-activated recording is used to reduce unnecessary footage, then storage space is conserved, but false motion detection from weather events and outdoor lighting activates recording for uninteresting clips
Solution Approach 1:
The patent changes the temporal parameter (frame rate) of recorded footage based on motion event priority. By playing back low-priority footage (such as weather-related motion) at faster frame rates during review, the system effectively filters out uninteresting content while maintaining accurate motion detection capabilities for future recording decisions.
Data Source
AI summary
A method for generating time-lapse representation of video footage in a security/automation system is described. In one embodiment, the method may include receiving video from a camera at a location associated with a home automation system, determining a first motion event in the video, identifying a level of priority associated with the first motion event, and generating a time-lapse representation of the video having a first frame rate for a first part of the time-lapse representation of the video and a second frame rate different from the first frame rate for a second part of the time-lapse representation of the video that includes the first motion event. In some cases, the time-lapse representation of the video may be generated based at least in part on the level of priority associated with the first motion event.