YOLOv11n & DeepSeek 1.5B LLM—Running Locally
Detect objects in a video stream
Identify and track faces in videos
Generate a video with stick figures tracking human poses
Detect objects and track body movements in real-time
Product Prototype 1
Video captioning/tracking
Track people in a video and capture faces
Detect objects in live video from your webcam
Detect objects in images, videos, and live webcam streams
Watch and analyze videos with object detection
Video captioning/open-vocabulary/zero-shot
Powerful foundation model for zero-shot object tracking
ThreatDetection is an advanced AI-powered tool designed for object detection and analysis in videos or CCTV streams. Leveraging cutting-edge technologies like YOLOv7 and DeepSeek 1.5B LLM, this application enables real-time tracking and identification of objects within video footage. It is optimized to run locally, ensuring privacy and efficiency for security and surveillance applications.
What is the minimum system requirement to run ThreatDetection?
ThreatDetection requires a capable GPU to handle real-time object detection. A minimum of 4GB VRAM is recommended for smooth performance.
Can I customize the types of objects ThreatDetection detects?
Yes, ThreatDetection allows you to define custom object categories and set specific detection parameters for improved accuracy.
Is ThreatDetection suitable for real-time CCTV monitoring?
Absolutely. ThreatDetection is designed for real-time analysis and can handle live CCTV streams effectively with low latency.