HOLa: HoloLens Object Labeling
*, Serouj Khajarian, Konstantin Holzapfel, Johannes Schmidt, Stefanie Remmele · 2024
Read HOLa: HoloLens Object Labeling by *, Serouj Khajarian, Konstantin Holzapfel, Johannes Schmidt, Stefanie Remmele (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-04945.txt.
Description
Labeling objects in augmented reality video is slow by hand, and a 2024 paper by Michael Schwimmbeck and co-authors offers a tool that does it almost alone. Their app, HOLa, runs on the HoloLens 2 headset and serves medical AR work, where tracking objects needs many precise masks.
HOLa builds on a tracking version of a leading segmentation model, so a user gives a light starting cue and the system follows the object through the video. It needs no tuning for a new scene's look. In tests on open liver surgery footage and lab phantom scenes, labeling ran more than five hundred times faster than by hand. Mask quality scores matched human annotators closely.
Catalog record
| CREATOR | *, Serouj Khajarian, Konstantin Holzapfel, Johannes Schmidt, Stefanie Remmele |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 16 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-04945