TRIT-Net: Triplet-based Railway Instance Tracing Network Using Attraction Field Representation
Bibliographic record
Abstract
Driverless trains with Grade of Automation 4 (GoA4) are reliably implemented in metro systems but lack local situational awareness, requiring human intervention during safety-critical events. In contrast, autonomous trains must perceive their environment and make real-time decisions similar to a human operator responses. This study focuses on rail route identification from RGB images, a key aspect of visual intelligence for autonomous railway. While existing methods have successfully achieved rail detection and segmentation, they struggle to generalize rail associations for route formation due to relying on hard-coded heuristics tailored to specific datasets. To address this, we propose TRIT-Net that simultaneously predicts left-right rail associations and route instances. Our architecture integrates convolutional layers with transformers, featuring a rail triplet regression branch and an Attraction Field Map (AFM) for instance-based route tracing, inspired by trajectory tracking theory in closed-loop linear control systems. Our experiments, ablation study, and stability analysis confirm the effectiveness of our approach in achieving comparatively higher precision, recall, and inference speed. We also demonstrate that our versatile model selection strategy is applicable to any off-the-shelf semantic segmentation baseline. Future work will focus on recognizing track switch configuration to extract a unique train route rather than multiple ego-path candidates.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".