Argus Yolov12 Crack

--- license: mit library_name: ultralytics pipeline_tag: object-detection tags: - yolov12 - crack-detection - wall-crack - object-detection - attention-mechanism - ema-attention - ablation-study datasets: - athena-cheung-sftri/wall-crack-3 language: en metrics: - mAP50 --- # Argus YOLOv12 Crack — Pretrained Weights This repository hosts the pretrained weights for **Argus YOLOv12 Crack**, an undergraduate-thesis-stage research project on wall crack detection. Recommended configuration: **`YOLOv12n + EMA(P3+P4, factor=16)`**. Code, training configs, reproduction guide, and Web demo: **https://github.com/Shallow-dusty/Argus-YOLOv12-Crack** ## Layout ``` . ├── kaggle/ # Paper-headline recommended weights (Kaggle P100, n=3) │ ├── ema-p34-factor16-seed0.pt │ ├── ema-p34-factor16-seed1.pt │ └── ema-p34-factor16-seed43.pt ├── 3070/ # Cross-platform check (RTX 3070 Laptop, n=3) │ ├── ema-p34-factor16-seed0.pt │ ├── ema-p34-factor16-seed1.pt │ └── ema-p34-factor16-seed43.pt ├── cross-domain/ # T6 cross-domain pressure test (single seed each) │ ├── ema-p34-factor16-crack500.pt │ └── ema-p34-factor16-deepcrack.pt └── baseline/ # YOLOv12n baseline (no attention) for reference ├── yolov12n-baseline-kaggle.pt └── yolov12n-baseline-3070-seed0.pt ``` ## Headline numbers Primary dataset: `wall_crack_v2_split` (2366 / 295 / 297, single class `crack`, CC BY 4.0). | Setting | Configuration | mAP50 | |---|---|:---:| | **Recommended · Kaggle n=3 mean ± std** | `YOLOv12n + EMA(P3+P4, factor=16)` | **0.73070 ± 0.00672** | | Recommended · RTX 3070 n=3 mean ± std | same | 0.73689 ± 0.00400 | | Baseline | `YOLOv12n baseline` | 0.70561 | Full data: see [`docs/COMPARISON_TABLE.md`](https://github.com/Shallow-dusty/Argus-YOLOv12-Crack/blob/main/docs/COMPARISON_TABLE.md) in the GitHub repo. ## Quick load (Ultralytics) ```python from huggingface_hub import hf_hub_download from ultralytics import YOLO weights = hf_hub_download( repo_id="ShallowDusty/argus-yolov12-crack", filename="kaggle/ema-p34-factor16-seed0.pt", ) model = YOLO(weights) results = model.predict("path/to/your/wall_image.jpg") ``` ## Training recipe (summary) - Backbone: YOLOv12n + EMA attention inserted at P3 + P4 - EMA factor (channel grouping divisor): 16 - Epochs: 200 with `patience=50` early stopping - Image size: 640 - Optimizer: SGD, lr0=0.01, momentum=0.937, weight_decay=5e-4 - Batch: 16...