Skip to content

Core Workflows

Core workflows are the primary tools researchers use to accomplish end-to-end tasks in primate face analysis.

Overview

These modules provide complete, integrated workflows for common research tasks:

  • Demos - Complete inference pipeline for face detection and pose estimation
  • DINOv2 - Feature extraction, visualization, and intelligent subset selection
  • GUI - Interactive annotation and pseudo-labeling workflow
  • Landmark Converter - Training and applying format conversion models

Workflow Integration

These modules are designed to work together:

  1. Data Collection → Use GUI for annotation
  2. Feature Analysis → Use DINOv2 for exploration
  3. Model Inference → Use Demos for detection/pose
  4. Format Conversion → Use Landmark Converter for compatibility

Getting Started

Each module has its own quick start guide. Start with the module that matches your immediate need:

  • New to PrimateFace? → Start with Demos
  • Need to analyze image features? → Use DINOv2
  • Have unlabeled data? → Try GUI
  • Need format compatibility? → Check Landmark Converter

For detailed API documentation, see the API Reference.