Human Extractor Skill

Description

GPU-accelerated pipeline for detecting, tracking, and classifying humans in dashcam footage. Processes MP4 videos to extract human crops with optional CLIP-based head covering classification, saving all outputs to a unified directory with comprehensive indexing.

Purpose

Extract visual evidence of human presence from dashcam recordings for investigative analysis. Optimized for high throughput using NVDEC decoding, batched YOLOv8 detection, ByteTrack multi-object tracking, and optional CLIP classification.

Usage

Basic Invocation

Extract humans from Park_R videos on October 6, 2025