Method on Efficient Operation of Multiple Models for Vision‐Based In‐Flight Risky Behavior Recognition in UAM Safety and Security
Bibliographic record
Abstract
The rapid development of urban air mobility (UAM) has emphasized the need for in‐flight control and passenger safety management. Recently, with the significant spread of technology in the field of computer vision, research has been conducted to manage passenger safety and security with vision‐based approaches. Previous research predominantly focuses on single‐task vision models, which limits their ability to comprehensively recognize various situations. In addition, conventional vision‐based deep learning models require substantial computational power, potentially reducing the operational sustainability of UAMs with limited electrical resources. In this study, we propose a novel cabin surveillance framework for passenger safety and security. The proposed method achieves high accuracy by using a single model optimized for a specific task and ensures maximum computational efficiency through a scheduler that executes the appropriate models based on the situation. It can effectively perform roles such as detecting prohibited items and recognition of dangerous/abnormal behavior. Moreover, it simplifies the management of the involved models by adding new models or updating the existing ones, and it provides a sustainable system by reducing energy consumption. Through comprehensive experiments on various benchmarks, we validated the effectiveness of each model and verified the practicality of the proposed framework in terms of time complexity and resource usage through practical tests.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".