Make-Sense

makesense.ai is a free online tool that lets you label and annotate photos for machine learning projects without installing anything—just open it in your browser and start working. If you're building an AI model that needs to recognize objects or people in images, you first need thousands of labeled examples. This tool makes that tedious process much faster and easier.

The way it works is straightforward: you upload your photos, then draw boxes around objects, mark key points on people's bodies, or trace outlines of shapes. You can label each region with a category (like "car" or "person"). The basic version lets you do all this manually, but there's also AI assistance built in. The tool can automatically detect objects in your photos and suggest where bounding boxes should go, or estimate where a person's joints are located, so you only need to correct or refine what it guesses rather than starting from scratch. All of this AI processing happens on your own device—your photos never get sent to a server, which means your data stays private.

Once you're done labeling, you can download your annotations in several standard formats (like COCO JSON, YOLO, or XML) that work with popular machine learning frameworks. This is useful if you're a student learning computer vision, a startup building a custom object detection model, or anyone preparing training data for a deep learning project. The tool runs in TypeScript and React, so it's built to work smoothly on Mac, Windows, and Linux without any complicated setup.

The README also includes keyboard shortcuts to speed up your workflow, the ability to import existing labels to build on previous work, and active development with plans to add more AI models in the future. If you need a simple, free, and privacy-conscious way to prepare image datasets for machine learning, this is a solid option.