v1.1.0

CVHub520/X-AnyLabelingv1.1.0Nov 6, 2023by CVHub520

AI Summary

This release adds pose estimation capabilities using YOLOv8-Pose, introduces object-level tagging support with yolov5_ram, enables adjustment of image preprocessing parameters (keep_prev_brightness and keep_prev_contrast), and adds batch labeling for unknown categories using Grounding-DINO.

Key Highlights

Added pose estimation support via YOLOv8-Pose
Enabled object-level tagging with yolov5_ram model
Added ability to adjust keep_prev_brightness and keep_prev_contrast parameters
Introduced batch labeling for arbitrary unknown categories using Grounding-DINO
Released new executables for CPU and GPU versions on Windows and Linux

New Features

YOLOv8-Pose support for pose estimation tasks
Object-level tag support with yolov5_ram
Adjustable keep_prev_brightness and keep_prev_contrast settings
Batch labeling capability for unknown categories based on Grounding-DINO

Full Release Notes

## New versions

- X-AnyLabeling-CPU.exe [md5 | d6c309095c260df5845fdd66d7a9039f]
- X-AnyLabeling-GPU.exe [md5 | d6ddd6f28b69a077555b21ea90fa15c1]
- X-Anylabeling-Linux-CPU [md5 | 68dff2464337bcfa3a14bcb36567343e]
- X-Anylabeling-Linux-GPU [md5 | 81785eb64605003d4bdf9934c98e97c4]

## New feature
- Support pose estimation: [YOLOv8-Pose](https://github.com/ultralytics/ultralytics). (#103 )
- Support object-level tag with yolov5_ram.
- Add the capability to adjust `keep_prev_brightness` and `keep_prev_contrast`. (#104 )
- Add a new feature enabling batch labeling for arbitrary unknown categories based on Grounding-DINO.