TULIQ IMAGE ANNOTATOR — OFFLINE EDITION
========================================

What this is
-------------
A bounding-box / shape annotation tool for building object-detection and
segmentation datasets — draw rectangles, circles, ellipses, polygons,
polylines, or points on your images, assign a class to each, and export in
YOLO, COCO, Pascal VOC, or VIA CSV format.

This is a fully offline version of the tool at https://www.tuliq.app/image-annotator.
It is a plain folder of files — no installer, no account, no internet
connection needed at any point after you've downloaded it.


Requirements
-------------
A modern desktop browser: Google Chrome, Microsoft Edge, or Mozilla Firefox.
That's it. No Node.js, no npm, no command line, nothing to install.


How to open it
----------------
Double-click "index.html". It opens directly in your default browser.
(If your browser is set to something unusual, right-click index.html,
choose "Open with", and pick Chrome, Edge, or Firefox.)


How to use it
---------------
1. Add your images — click "Add images" to pick individual files, or
   "Add a folder" to load every image in a folder at once.
2. Add at least one class name in the "Classes" panel on the right.
3. Pick a shape from the toolbar above the image (rectangle, circle,
   ellipse, polygon, polyline, or point) and draw on the image:
     - Rectangle / circle / ellipse: click and drag.
     - Polygon / polyline: click to add each point, then double-click
       (or press Enter) to finish, or press Esc to cancel.
     - Point: click once to place it.
4. Click any existing box to select it — drag its handles to resize it,
   or drag the shape itself to move it. Press Delete or Backspace to
   remove the selected box.
5. Keyboard shortcuts: number keys 1-9 switch your active class,
   Left/Right arrow keys switch images.
6. When you're done, pick an export format and click "Export". A single
   file downloads directly; multiple files download as a .zip — both go
   to your browser's normal Downloads folder, exactly like any other
   file you'd download.

IMPORTANT: this offline version does not autosave. Export your work
before closing the tab if you want to keep it — closing or refreshing
loses whatever hasn't been exported yet.


Export formats
-----------------
YOLO        One .txt file per image plus a classes.txt — normalized 0-1
            coordinates, bounding boxes only.
COCO        A single combined annotations.json covering every image —
            bounding boxes only.
Pascal VOC  One .xml file per image — bounding boxes only.
VIA CSV     A single combined annotations.csv — the only format here that
            preserves every shape (circle, ellipse, polygon, polyline,
            point) exactly as drawn, not just its bounding box.

Circles, ellipses, and polygons flatten to their bounding box for
YOLO/COCO/Pascal VOC, since those formats only understand rectangles.
Points and polylines have no meaningful bounding box, so they're skipped
for those three formats — you'll see a note if anything was skipped.


Privacy
---------
Everything happens on your own computer. There is no server involved at
all — not even a website. Your images and annotations are never uploaded,
logged, or transmitted anywhere, at any point.


Folder contents
------------------
index.html    The app itself — open this file.
app.js        All of the app's logic (plain JavaScript, no build step).
styles.css    Visual styling.
assets/       A vendored copy of JSZip (used to bundle multi-file exports
              into a .zip), included under its MIT license — see
              assets/jszip-LICENSE.txt.
README.txt    This file.


More tools
------------
This is one tool from the Tuliq toolset — small, fast, browser-first
utilities for developers, geospatial analysts, and technical creators.
See the rest at https://www.tuliq.app (an internet connection is only
needed to visit that site, never to use this offline tool).
