Redefining Real-Time Road Quality Analysis With Vision Transformers on Edge Devices
Bibliographic record
Abstract
Road infrastructure is essential for transportation safety and efficiency. However, the current methods for assessing road conditions, crucial for effective planning and maintenance, suffer from high costs, time-intensive procedures, infrequent data collection, and limited real-time capabilities. This paper presents an efficient lightweight system to analyze road quality from video feeds in real-time. The backbone of the system is EdgeFusionViT, a novel vision transformer (ViT)-based architecture that uses an attention-based late fusion mechanism. The proposed architecture outperforms lightweight CNN-based and ViT-based models. Its practicality is demonstrated by its deployment on an edge device, the Nvidia Jetson Orin Nano, enabling real-time road analysis at 12 frames per second. EdgeFusionViT outperforms existing benchmarks, achieving an impressive accuracy of 89.76% on the Road Surface Condition Dataset (RSCD). Notably, the model maintains a commendable accuracy of 76.89% even when trained with only 2% of the dataset, demonstrating its robustness and efficiency. These findings highlight the system’s potential in road infrastructure management. It aids in creating safer, more efficient transport systems through timely, accurate road condition assessments. The study sets a new benchmark and opens up possibilities for advanced machine learning in infrastructure management.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 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".