An IIIF explorer for high-resolution images, audio, video, and 3D models in a modern, performance-focused interface.
Mimir is being developed as part of digital.ub.uni-koeln.de. The current state should be understood as work in progress: some features are still missing, parts of the interaction model are still being refined, and performance improvements remain an active area of development. Our goal is to reach a stable 1.0 version in 2026.
If a manifest is part of a collection, Mimir can display the other manifests within that collection in a way that keeps navigation continuous. In a presentation setting, this helps move between related objects without breaking the surrounding context.
MIMIR places the focus on full-text highlighting and the presentation of OCR-derived text in relation to the document image. Text and image remain linked, colors can be adjusted to suit different devices and viewing conditions, and the text view can be switched between running text and line-based transcription.
MIMIR focuses on selecting and addressing a spatial region within a canvas. The surrounding regions can be visually de-emphasised, and the blurring of those areas can be toggled depending on whether the presentation should isolate or recontextualise the selected area.
The reading interface can switch between different behaviours, such as book mode and continuous reading. At the same time, the authors intended reading behaviour remains preselected, so the presentation can begin from the mode that best reflects the structure of the source.
MIMIR focuses on audio and video material as part of a broader IIIF presentation context. The interface remains close to the image-based views, which can be useful when demonstrating mixed collections or teaching across media types.
MIMIR isolates the 3D view as its own demo situation. It suggests how spatial objects may be presented alongside other IIIF resources while preserving a recognisable interface structure.