Skip to content

Releases: seba2390/Bbannotate

v1.1.0

14 Feb 20:16
950a886

Choose a tag to compare

What's Changed

Full Changelog: v1.0.9...v1.1.0

v1.0.9

14 Feb 12:55
fc8aaf5

Choose a tag to compare

Added

  • Adjustable draw crosshair controls in canvas toolbar (arm length and stroke width)
  • Persistent crosshair preferences via localStorage

Changed

  • Refined the canvas top toolbar into compact titled groups (Controls, Zoom, Label, Box, Cursor) with consistent control heights
  • Updated Controls to always-visible compact tool buttons (Select, Draw, Pan) under the title
  • Moved zoom actions into a compact expandable panel under Zoom to reduce toolbar width
  • Replaced always-visible cursor sliders with compact expandable controls for arm length and arm width to keep toolbar footprint smaller
  • Increased maximum crosshair arm length from 48 to 96 for longer cursor arms
  • Switched draw-mode crosshair rendering to a canvas overlay to keep long cursor arms stable at larger sizes
  • Lowered minimum crosshair arm width from 1.0 to 0.5 for finer cursor thickness tuning

Fixed

  • Prevent drawing from starting outside image bounds
  • Constrain in-progress drawing to image boundaries (including auto-pan updates), so boxes cannot be dragged past image edges
  • Fixed crosshair shrinking at larger arm lengths due browser cursor image-size limits
  • Center images by default in the canvas viewport (including zoom reset)

v1.0.8

10 Feb 23:11

Choose a tag to compare

Fixed a cli startup issue

v1.0.7

10 Feb 21:38

Choose a tag to compare

Bbannotate Release Notes

This release improves annotation clarity, runtime lifecycle behavior, and CLI observability, with a strong focus on responsiveness.

Highlights

  • Added a new bounding-box color tool in the image toolbar:
    • Auto mode chooses a high-contrast color against the underlying image region.
    • Label mode keeps label-based colors.
    • Custom mode allows manual color selection.
  • Bounding box color now applies while drawing, not only after the box is finalized.
  • Improved selected-box visibility:
    • Clearer highlight on canvas.
    • Stronger visual emphasis for selected items in the right-side annotation list.

Runtime & Process Management

  • bbannotate start now runs detached by default (except reload mode), so closing the terminal does not interrupt the app.
  • Added browser-session lifecycle handling so closing the browser window cleanly shuts down related running processes.
  • Added bbannotate status to inspect current runtime state (running services/processes and health status).

Documentation

  • README updated to reflect:
    • New color tooling behavior.
    • Startup/runtime modes.
    • bbannotate status usage.
    • Current command behavior and shortcuts.

Notes

  • No breaking workflow changes expected for normal use.

v1.0.6

07 Feb 20:34
a67d573

Choose a tag to compare

What's Changed

Full Changelog: v1.0.5...v1.0.6

Links

v1.0.5

05 Feb 22:08
0e3eec1

Choose a tag to compare

bbannotate v1.0.5

  • Bug fix
  • Design update

Links

v1.0.4

05 Feb 20:44
a0288f7

Choose a tag to compare

bbannotate v1.0.4

A lot of bug fixes
Updates design

Links

v1.0.3

04 Feb 22:31

Choose a tag to compare

Fixed

  • Wait for server to be ready before opening browser (prevents ERR_CONNECTION_REFUSED)
  • Fixed Windows test timing issue for project ordering
  • Fixed Makefile to build frontend before Python install

v1.0.2

04 Feb 22:29

Choose a tag to compare

Fixed

  • Fixed CI workflow to build wheel directly (avoiding sdist which excludes frontend/dist)
  • Fixed CLI help test to handle Rich ANSI color codes in output
  • Fixed .gitignore to properly track frontend/src/lib/api.ts

v1.0.0

04 Feb 20:57

Choose a tag to compare

🎉 Initial Release

bbannotate - A lightweight bounding box annotation tool for image datasets.

Installation

pip install bbannotate
bbannotate start

Features

  • 🖼️ Multi-format support — PNG, JPEG, WebP, BMP
  • 📁 Project management — Organize annotations by project
  • 🏷️ Custom labels — Define your own class labels
  • ⌨️ Keyboard shortcuts — Fast annotation workflow
  • 📤 Multiple export formats — YOLO, COCO, Pascal VOC, CreateML, CSV
  • 🔄 Train/Val/Test split — Automatic dataset splitting for YOLO export

Links