ai-kissing

A Rust crate providing utilities for detecting and analyzing kissing events in video streams. Designed for research and development in areas like behavioral analysis and human-computer interaction.

Installation

To use ai-kissing in your Rust project, add the following to your Cargo.toml file: toml [dependencies] ai-kissing = "0.1.0" # Replace with the latest version

Usage

This crate provides functionalities to process video frames, identify potential kissing events based on proximity and facial feature analysis, and generate reports.

Example 1: Basic Kissing Event Detection rust use ai_kissing::{FrameData, KissingDetector};

fn main() { // Simulate frame data (replace with actual video frame processing) let frame1 = FrameData { person1_x: 100, person1_y: 150, person2_x: 110, person2_y: 160, // ... other relevant data like facial landmarks };

let frame2 = FrameData {
    person1_x: 105,
    person1_y: 155,
    person2_x: 108,
    person2_y: 158,
    // ... other relevant data like facial landmarks
};

let mut detector = KissingDetector::new();

if detector.is_kissing(&frame1) {
    println!("Potential kissing event detected in frame 1!");
}

if detector.is_kissing(&frame2) {
    println!("Potential kissing event detected in frame 2!");
}

}

Example 2: Tracking Kissing Duration rust use ai_kissing::{FrameData, KissingDetector};

fn main() { let mut detector = KissingDetector::new(); let mut kissing_start_time = None;

// Simulate video frames
let frames = vec![
    FrameData { person1_x: 50, person1_y: 60, person2_x: 52, person2_y: 62 },
    FrameData { person1_x: 51, person1_y: 61, person2_x: 51, person2_y: 61 },
    FrameData { person1_x: 52, person1_y: 62, person2_x: 50, person2_y: 60 },
    FrameData { person1_x: 70, person1_y: 80, person2_x: 90, person2_y: 100 },
];

for (index, frame) in frames.iter().enumerate() {
    if detector.is_kissing(frame) {
        if kissing_start_time.is_none() {
            kissing_start_time = Some(index);
            println!("Kissing started at frame: {}", index);
        }
    } else {
        if kissing_start_time.is_some() {
            let start = kissing_start_time.unwrap();
            println!("Kissing ended at frame: {}.  Duration: {} frames", index, index - start);
            kissing_start_time = None;
        }
    }
}

}

Example 3: Customizing Kissing Detection Thresholds rust use ai_kissing::{FrameData, KissingDetector, DetectorConfig};

fn main() { let config = DetectorConfig { proximity_threshold: 15, // Increased proximity threshold facial_feature_similarity_threshold: 0.8, // Changed feature similarity threshold };

let mut detector = KissingDetector::with_config(config);

// Simulate frame data
let frame = FrameData { person1_x: 20, person1_y: 30, person2_x: 34, person2_y: 40 };

if detector.is_kissing(&frame) {
    println!("Kissing event detected with custom configuration!");
}

}

Features

• Kissing Event Detection: Identifies potential kissing events based on proximity and facial feature analysis.
• Configurable Parameters: Allows customization of detection thresholds for proximity and facial feature similarity.
• Frame-by-Frame Analysis: Processes video frames to track kissing events over time.
• Duration Tracking: Calculates the duration of kissing events in terms of frames.
• Simple API: Easy-to-use functions for integrating kissing detection into your video processing pipelines.

License

MIT

This crate is part of the ai-kissing ecosystem. For advanced features and enterprise-grade tools, visit: https://supermaker.ai/video/ai-kissing/