Image annotation and object detection tool
LCFinder (LC's Finder) is an image management tool that supports image annotation and object detection. It is written in C and uses LCUI for its graphical interface. As with the author's other projects, the naming is simple, it begins with LC, and the following Finder is referenced from the Finder in Mac OS.
LCFinder's user interface and feature design is based on the "photos" application that comes with Windows, Although it is a reference, the functional aspects are mainly developed according to the author's individual needs, and the author does not intend to waste time to implement all the functions of the "photo" application.
- Image annotation: Provides a simple GUI for marking bounded boxes of objects in images for training Yolo v3 and v2
- Object detection: Built-in image detector, It can automatically annotate the detected objects in the images.
- Search by tags: You can browse and search tagged images in the tags view
- Localization: Support English, Simplified Chinese, Traditional Chinese, expandable support for other languages.
- Private space: A password-protected space where you can hide non-public image sources
- UWP: Support for Windows Universal Platform (UWP), you can click this link to view it in the windows app store. Due to the development cost of the UWP version, it will not be updated with the desktop version.
If you want to use the detector, you need the following steps:
- Download pre-trained models:
yolov3.cfg(236 MB COCO Yolo v3) - requires 4 GB GPU-RAM: https://pjreddie.com/media/files/yolov3.weights
yolov3-tiny.cfg(34 MB COCO Yolo v3 tiny) - requires 1 GB GPU-RAM: https://pjreddie.com/media/files/yolov3-tiny.weights
yolo9000.cfg(186 MB Yolo9000-model) - requires 4 GB GPU-RAM: http://pjreddie.com/media/files/yolo9000.weights
- Copy the
.weightsfile to its namesake directory in the
app/detector/modelsdirectory, such as: copy
If you are interested in fixing issues and contributing directly to the code base, please see the contributing guidelines, which covers the following:
- How to build and run from source
- Project Structure
- Submitting pull requests
- Contributing to translations
The development of LCFinder is inseparable from the support of these projects:
- LCUI — UI engine, provide graphical user interface support
- LCUI.css — UI component library, provides basic styles and components for the graphical user interface
- darknetlib — C bindings for darknet, provide image recognition support
Licensed under the GPL License.