Leveraging Anatomical Guidance: A Novel Attention Mechanism for Pedestrian Detection
Bibliographic record
Abstract
Effective pedestrian detection requires robust feature representations adaptable to varying human appearances. While deep convolutional networks dominate the field, they often lack explicit anatomical priors. This paper introduces Pedestrian Anatomy-Aware (PAA) Attention, a novel, lightweight, and interpretable attention module designed to address this limitation. PAA employs an anatomically inspired structural attention mechanism with parallel vertical and horizontal kernels to capture essential structural cues. It further enhances feature representation through spatial and channel attention, complemented by an identity gate for stable integration. Designed as a general plug-and-play module, PAA is evaluated within the YOLOv8s object detection framework. Experiments on the CityPersons dataset show that YOLOv8s integrated with PAA achieves significant improvements in pedestrian detection, yielding 4.4% and 3.0% increases in mAP50 and mAP50-90, particularly benefiting occluded and small pedestrians. Furthermore, attention map visualizations confirm that PAA enhances interpretability by directing the model’s focus toward anatomically relevant regions. The modular design of PAA provides a broadly applicable strategy for incorporating anatomical awareness into diverse object detection frameworks, offering a promising direction for advancing vision applications.
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) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| 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".