MRLabeling: Create RGB-D Datasets On-The-Fly using Mixed Reality
Bibliographic record
Abstract
One of the best ways to build better vision models is to train large models on big datasets. However, the process of building such datasets is often costly and tedious. With the ever increasing adoption of Mixed Reality in professional settings, and with the performance improvements of headsets in recent years, we see an opportunity for a tool that combines data collection and annotation in a single process, and that leverages both RGB and depth data provided by the headset sensors. Moreover, assisting machine learning predictive models with user inputs through natural mixed reality interactions is a promising prospective for human-artificial intelligence interactions. In this paper, we present MRLabeling, an application developped for the Microsoft Hololens 2 that allows the easy creation and annotation of datasets directly in Mixed Reality. We first describe the design of the system, the way 3D bounding boxes drawn by the user are projected in 2D to create annotated images on the fly, and the use of segmentation algorithms to go beyond bounding boxes. After that, we explore the use of depth data, and the current limitations of the system, as well as avenues for future work.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.014 | 0.010 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".