temporal-persistence-event

Detect sustained occupancy or anomalies in video frames using timestamp analysis.

Updated Jun 2, 2026

Search Tags:#video processing #anomaly detection #temporal analysis #event detection #video surveillance

Ready to run

npx skills add https://github.com/bettercallfan/deerflow --skill temporal-persistence-event

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: temporal-persistence-event
Source: https://github.com/bettercallfan/deerflow/tree/main/skills/custom/video_surveillance/temporal-persistence-event
Command: npx skills add https://github.com/bettercallfan/deerflow --skill temporal-persistence-event

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill identifies whether an object, action, or scene state in a video persists long enough to be considered a sustained event or anomaly.

Core Features & Use Cases

Event Detection: Detects events like lingering, illegal parking, fallen person, congestion, smoke/fire, blockage, or intrusion.
Visual Analysis: Uses extracted frames for visual analysis and compares timestamps to determine persistence.
Duration Estimation: Estimates the duration of events based on visible evidence timestamps.
Use Case: For instance, monitoring a parking lot for illegal parking or identifying a fallen person in a public area.

Quick Start

Run the temporal-persistence-event skill with the provided frame data and target labels to detect events of interest.

Frequently Asked Questions about temporal-persistence-event

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I detect sustained events in video surveillance footage?▼

Event persistence in video sequences is determined by analyzing extracted frames and comparing their timestamps to estimate duration, confirming whether visual evidence of an anomaly or occupancy lasts long enough to trigger an alert.

What types of anomalies can temporal video analysis identify?▼

Temporal video analysis can identify anomalies like lingering, illegal parking, fallen persons, congestion, smoke or fire, blockage, and intrusion by verifying if the scene state persists across multiple extracted frames.

Do I need to extract frames before running video anomaly detection?▼

Yes, the skill requires extracted frames as input. It operates on these extracted frames and target labels to perform visual analysis and estimate the duration of detected events.

How does temporal analysis estimate the duration of a detected event?▼

Temporal analysis estimates event duration by comparing the timestamps of visible evidence across extracted frames, calculating how long an object or scene state persists to confirm it meets sustained occupancy thresholds.

Can I use this for monitoring illegal parking in a parking lot?▼

Yes, you can use this skill for monitoring illegal parking in a parking lot. It detects sustained occupancy by verifying if a vehicle persists in a specific location across extracted frame timestamps.

What are the limitations of frame-based temporal event detection?▼

Frame-based temporal event detection relies on extracted frames rather than continuous video streams, meaning its duration estimation accuracy depends on the frame extraction rate and the visual evidence available in those discrete timestamps.