🚀 We’re excited to announce a major update to PixLab Annotate, our powerful browser-based image annotation and segmentation tool used by developers, researchers, and ML teams worldwide. With this release, Annotate is faster, more intuitive, and packed with new features designed to streamline dataset preparation for your machine learning projects.
🔗 Try it now at: annotate.pixlab.io
🆕 What’s New in This Update?
🗂️ Improved Bulk Uploading & Management
Handling large datasets just got easier: - Drag-and-drop multiple images or full folders - Real-time progress tracking - Faster loading and rendering of high-resolution images
✏️ Upgraded Annotation Tools
We’ve added more flexibility and precision to our tools: - Polygon and rectangle tools are now more responsive and support fine-grain editing - New free-draw segmentation mode for pixel-perfect outlines - Snap-to-edge enhancements for cleaner boundaries
📤 Export Enhancements
- Export all annotations in a single, structured JSON file
- Supports popular formats for TensorFlow, PyTorch, Detectron2, and others
- Optional image mask export (PNG) for segmentation tasks
🎯 Label System Overhaul
Labeling is now more powerful and organized: - Brand-new Label Manager with color-coding, grouping, and edit capabilities - Quickly switch between labels using keyboard shortcuts - Assign and auto-fill labels for repetitive tasks
⏪ Full Annotation History + Auto-Save
- Undo/redo every action with confidence
- Auto-save to local storage so you never lose work
- Load previous sessions instantly from your browser
🧩 Integration with PixLab Tools Ecosystem
- Seamless bridge to the Vision Playground for testing AI models on your annotated data
- Direct compatibility with PixLab’s API for automated processing, content moderation, and more
🔐 Privacy-First, Always
Like before, all annotations are stored locally in your browser. No uploads. No tracking. You’re in full control of your data.
⚡ Why This Matters
This update transforms Annotate into a production-ready tool for: - Preparing custom training datasets - Labeling at scale with higher accuracy - Rapid prototyping and iteration cycles in ML workflows
Whether you're training a YOLO model, segmenting for instance detection, or prepping for OCR, PixLab Annotate now meets the needs of both rapid prototyping and large-scale data labeling.
🚀 Ready to Explore?
Head over to annotate.pixlab.io and experience the upgrade firsthand.
We’d love your feedback! More features — including S3 export, collaborative annotation, and multi-user workspaces — are on the way. Stay tuned.
Happy annotating!