​
Computer Vision Without the Complexity

PixelFlow transforms how you work with computer vision. One library, any framework, maximum performance. Built for developers who need CV to work reliably across different models and environments.

Extreme Simplicity

One function does one thing perfectly. No complex abstractions or steep learning curves.

Extreme Performance

OpenCV-optimized drawing primitives built for production-speed processing.

Complete Reusability

Write once, use with YOLO, Detectron2, MediaPipe, or any CV framework.

​
The Simplest Computer Vision API

Here’s how simple it is: 
import pixelflow as pf
from ultralytics import YOLO

# Any YOLO model works
model = YOLO('yolov8n.pt')
results = model('image.jpg')

# Convert to PixelFlow format
detections = pf.from_ultralytics(results[0])

# Professional annotations in one line
annotated = pf.annotate.box(image, detections)

Get Started Now

Complete installation and first application in under 5 minutes

​
What You Can Build

PixelFlow works across industries - from retail analytics to manufacturing quality control, security monitoring to healthcare imaging. The unified interface means your code works regardless of the underlying model.

Object Detection & Tracking

Multi-object tracking with ByteTracker, zone-based filtering, and crossing detection for comprehensive scene analysis.

Visual Annotations

Professional-grade annotations including boxes, masks, keypoints, blur effects, and custom overlays.

Spatial Analytics

Zone management, line crossing detection, and spatial filtering for location-based insights.

Performance Tools

Built-in timing, FPS counters, and memory-efficient video processing for production deployments.

​
Framework Compatibility

PixelFlow provides unified converters for all major computer vision frameworks, eliminating vendor lock-in and enabling framework mixing within the same project.
Supported Frameworks:
  • YOLO/Ultralytics - Real-time object detection
  • Detectron2 - Research-grade instance segmentation
  • MediaPipe - Hand/pose/face detection
  • MMDetection - Advanced detection models
  • HuggingFace Transformers - Vision transformers
  • Custom Models - Build your own converters

​
Quick Navigation

Detections

Convert any model output to PixelFlow’s unified format

Annotations

Professional visual overlays and effects

Tracking

Multi-object tracking across video frames

Zones

Spatial filtering and region management
New to computer vision? Start with the Quickstart Guide for a complete walkthrough from installation to your first working application.