digiKam 8.3 Introduces Automatic Image Tagging with Deep Learning Neural Networks

digiKam 8.3 has been released, offering a range of new features and improvements. One of the key highlights is a new tool that utilizes a Deep Learning neural network engine to automatically tag images, recognizing objects, scenes, and events in digital photos.

Additionally, this release introduces a tool to apply metadata from images or JSON files to digital photos. It also includes a revamped video preview and slideshow, now rendered with Qt6::Multimedia and Qt5::QtAVPlayer, compatible with FFmpeg 5 and later. Other additions include a new settings page for customizing geolocation parameters and generic network proxy settings.

Older and unmaintained QtAV framework code has been removed, and the AppImage universal binary now uses Qt 5.15.12 LTS and KDE Frameworks 5.115. The LibRaw library, used for reading raw files from digital cameras, has been updated to snapshot 2024-02-02.

The new release also introduces support for the Imagga image recognition API and computer vision AI. It improves support for the Olympus C-310 digital camera on SUSE Linux systems and enhances image detection from Sanyo VPC-G200 cameras.

Various bugs have been fixed in this release, improving functionalities such as file deletion, renaming, face naming in image previews, sharing and loading maps from other mapping software, AF points in focus, support for XFC images created with GIMP 2.10.36, and face tagging.

For a detailed list of changes, you can refer to the changelog here. digiKam 8.3 is available for download as an AppImage, compatible with most GNU/Linux distributions, without the need for installation.

0 / 5

Your page rank:

Subscribe: YouTube page opens in new windowLinkedin page opens in new windowTelegram page opens in new window

Leave a Reply

Your email address will not be published. Required fields are marked *

Post comment