search

ryouchinsa / Rectlabel-support

RectLabel is an offline image annotation tool for object detection and segmentation.
https://rectlabel.com
498 stars 73 forks source link

Feature Feedback and Launch Inquiry for Windows and Linux systems #249

Open salwaghanim opened 5 months ago

salwaghanim commented 5 months ago

Dear Rectlabel support team

I'm writing to express my appreciation for your development of [Tool Name]. I've been evaluating several labeling tools, and yours quickly rose to the top. It offers all the essential features I look for in a labeling tool and provides a smooth, offline experience.

However, there's one aspect where Roboflow currently holds an edge: their tool includes health checks, statistical features, and even an auto-labeling tool powered by trained models. While these features would be beneficial additions to Rectlabel, your tool's core functionality is excellent.

On that note, I'm curious about the release roadmap for Rectlabel. Any chance you could offer insights into its potential launch on Ubuntu or Windows platforms?

Thank you for your time and consideration. I look forward to seeing how Rectlabel continues to evolve!

ryouchinsa commented 4 months ago

Thanks for your kind feedback.

This is our roadmap to evolve RectLabel. Please let us know your opinion.

  1. Correspond to ONNX models for auto labeling such as EfficientSAM, EdgeSAM, HQ-SAM, OpenMMLab, Detectron2, etc.
  2. Customize RectLabel to make it easy to use with Roboflow and FiftyOne.
  3. Develop the open source version for macOS.
  4. Develop the open source version for Ubuntu and Windows.

Update: Version 2024.03.03 corresponded to the EfficientSAM models and text recognition for words. sam-cpp-macos corresponded to EdgeSAM. To add to RectLabel apps, we are checking the license policy to the developer. Version 2024.03.13 corresponded to the HQ-SAM models.