roboflow/supervision
Supervision
Supervision is a computer vision toolset for normalizing model outputs, managing datasets, and visualizing annotations. It provides a framework to convert predictions from various classification and detection models into a standardized data format to ensure interoperability across different computer vision pipelines.
The library features a post-processor for filtering, counting, and tracking detected objects across image frames and video streams. It includes capabilities for large image tiling to improve the detection of small objects and tools for assigning persistent identities to objects to monitor movement across polygonal zones and boundary lines.
The project covers a broad range of computer vision utilities, including dataset format conversion, image geometry transformations, and the rendering of visual overlays such as bounding boxes and masks. It also provides tools for benchmarking model performance using standardized accuracy metrics and confusion matrices.
Features
- Computer Vision Post-Processing Toolsets - Offers a comprehensive collection of utilities for processing, normalizing, and visualizing detections from various computer vision models.
- Computer Vision Normalizers - Provides a framework to convert predictions from diverse computer vision models into a standardized data format.
- Bounding Box Visualizers - Draws bounding boxes, masks, and labels on images and video to visually verify model predictions.
- Standardized Detection Formats - Converts imagery processed through various computer vision models into a standardized detection format for interoperability.
- Post-Processing Utilities - Standardizes and filters predictions from different models to prepare visual data for analysis or business logic.
- Object Detection and Tracking - Assigns persistent IDs to detected objects across video frames to count items crossing specific lines or zones.
- Dataset Management - Provides tools to load, split, merge, and convert image datasets for training and evaluation purposes.
- Vision Dataset Standardizers - Converts computer vision datasets between different formats to ensure compatibility between training and evaluation frameworks.
- Detection Visualization - Offers tools for overlaying bounding boxes and landmarks on images and video to verify detection accuracy.
- Detection Visualization Libraries - Provides a suite of tools for overlaying bounding boxes, masks, and labels on imagery to visualize model results.
- Visual Annotation Tools - Provides a suite of tools for overlaying bounding boxes, masks, and labels onto images and video frames.
- Dataset Management Tools - Provides utilities for converting computer vision datasets between common formats to ensure model compatibility.
- Object Tracking - Assigns persistent identifiers to detected objects across video frames to maintain identity over time.
- Spatial Detection Zones - Tracks objects crossing specific lines or residing within polygon zones for spatial filtering and counting.
- Computer Vision Detection Formats - Provides utilities to convert model outputs from various computer vision frameworks into unified objects for simplified data handling.
- Data Normalization Utilities - Transforms detection results from different computer vision models into a unified internal object model for consistent processing.
- Vision Dataset Converters - Transforms computer vision datasets between different common formats to ensure compatibility between training and evaluation frameworks.
- Computer Vision Utilities - Provides helper scripts and utilities for processing and visualizing detection results in vision workflows.
- Model Performance Benchmarking - Evaluates computer vision model accuracy using standardized metrics like mean Average Precision and confusion matrices.
- Model API Integrations - Integrates outputs from classification, detection, and segmentation models using a consistent data format.
- Result Management - Ships tools to sort, count, and merge detection results by class or name to isolate specific targets.
- Image Slicing Pipelines - Divides high-resolution images into overlapping patches to improve detection accuracy for small objects.
- Computer Vision Pipelines - Standardizes the processing and visualization of detection results within computer vision pipelines.
- Image Transformation Utilities - Provides utilities to crop, scale, resize, and pad images to modify dimensions for model input.
- Media Stream Loaders - Includes utilities to read image files or iterate through video frames for processing pipelines.
- AI Tools - Computer vision toolkit for detection and segmentation.
- Vision par ordinateur - Reusable utilities for computer vision tasks.
- Computer Vision Frameworks - Tools for managing computer vision pipelines and monitoring.
