YOLO26s Finetuned on BDD100K Scenario Classification


Task Framework Base Model
Macro F1 Top-1 Params
License Source

Fine-tuned YOLO26s image classifier on the BDD100K Scenario Classification dataset, trained and evaluated as part of BDD100K-Toolkit, a dependency-clean toolkit for preparing BDD100K, training models on it and evaluating them with the same metrics on the same splits.

7-class driving-scenario classification (city street / highway / residential / parking lot / gas stations / tunnel / unknown) derived from BDD100K's per-image attributes.scene field. Unofficial task; follows the Kaggle dataset of the same name.


Usage

from huggingface_hub import hf_hub_download
from ultralytics import YOLO

model = YOLO(hf_hub_download("dronefreak/bdd100k-scenario-yolo26s-cls", "best.pt"))
probs = model("street.jpg", verbose=False)[0].probs
print(model.names[probs.top1], f"{float(probs.top1conf):.1%}")

Results

Evaluated on the test split (10000 images).

Metric Value
Macro F1 52.24%
Balanced accuracy 49.23%
Macro precision 58.28%
Macro recall 49.23%
Top-1 accuracy 77.81%
Top-5 accuracy 99.94%

Per class

Class Precision Recall F1 Test images
city street 79.14% 88.47% 83.54% 6112
gas stations 0.00% 0.00% 0.00% 7 (few)
highway 78.37% 67.99% 72.81% 2499
parking lot 55.32% 53.06% 54.17% 49 (few)
residential 68.16% 49.72% 57.50% 1253
tunnel 86.96% 74.07% 80.00% 27 (few)
unknown 40.00% 11.32% 17.65% 53

This checkpoint classifies none of the test images of gas stations (7 images) correctly, so treat it as unsupported.

Normalized confusion matrix


Model Zoo

All runs below were evaluated on the same test split, sorted by top-1 accuracy.

Model Top-1 Macro F1 Balanced acc Macro precision
EfficientViT-B3 79.32% 57.70% 53.59% 71.54%
YOLO11n 78.58% 49.47% 46.05% 60.98%
YOLO11s 78.58% 49.18% 45.24% 61.98%
MobileNetV4-Conv-Small 78.20% 52.89% 48.69% 61.41%
MobileNetV4-Conv-Large 78.15% 50.00% 45.48% 63.03%
EfficientViT-B1 78.01% 54.15% 49.41% 70.13%
TinyViT-21M 77.99% 62.52% 66.21% 59.66%
YOLO26s 77.81% 52.24% 49.23% 58.28%
ConvNeXt-Atto 77.44% 61.06% 60.34% 67.92%
YOLOv8s 77.15% 48.43% 46.64% 51.16%
ResNet-18 77.14% 47.29% 44.83% 54.25%
YOLO26n 76.78% 48.33% 43.94% 58.99%
ConvNeXt-Tiny 76.69% 56.88% 59.82% 54.60%
YOLOv8n 76.13% 46.40% 43.26% 54.14%
EfficientViT-B0 75.94% 46.10% 42.64% 53.74%

Training

Setting Value
Epochs (max) 50
Epochs (trained) 19
Best epoch (best.pt) 9
best.pt chosen by macro_f1 on the valid split
Early stopping patience 10
Batch size 128
Image size 224
Optimizer auto, resolved to MuSGD(lr=0.01, momentum=0.9)
Pretrained True
Seed 0

Dataset

dronefreak/BDD100K-Scenario-Classification holds the prepared splits these models were trained and evaluated on.


Limitations

  • Unofficial task: labels are BDD100K's per-image attributes, not a benchmark with a public leaderboard, so scores are only comparable with other models evaluated on this split.
  • Not the official test set: test here is BDD100K's official validation split (the official test labels are not released) and valid is a seeded 15% slice of the official train split.
  • Heavily imbalanced: rare classes have very few test images, so their per-class scores are noisy. Prefer macro F1 over accuracy.
  • Images are US dashcam frames; generalization to other regions or camera setups is untested.
  • BDD100K is released for non-commercial research and education. Check its license before any use of these weights beyond research.
  • scene is a coarse per-image label; frames showing several settings (e.g. a highway entering a tunnel) get a single class.

License

The weights are released under the license in the metadata above. They were trained on BDD100K, which is free for non-commercial research and education; commercial use needs separate permission (see https://www.bdd100k.com/). The prepared dataset on the Hub is tagged license: other, and the original BDD100K terms still apply.


Citation

Model

@article{jocher2026yolo26,
  title={Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models},
  author={Jocher, Glenn and Qiu, Jing and Liu, Mengyu and Lyu, Shuai and Akyon, Fatih Cagatay and Kalfaoglu, Muhammet Esat},
  journal={arXiv preprint arXiv:2606.03748},
  year={2026}
}

Dataset

@article{yu2018bdd100k,
  title={BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning},
  author={Yu, Fisher and Chen, Haofeng and Wang, Xin and Xian, Wenqi and Chen, Yingying and Liu, Fangchen and Madhavan, Vashisht and Darrell, Trevor},
  journal={arXiv preprint arXiv:1805.04687},
  year={2018}
}

Files

  • best.pt
  • metrics.json
  • results.csv
  • results.png
  • confusion_matrix.png
  • confusion_matrix_normalized.png
  • val_batch0_pred.jpg
  • assets/demo_banner.mp4
  • assets/demo_banner_poster.jpg
  • README.md

Reproduce

Trained and evaluated with BDD100K-Toolkit: bdd100k-evaluate --dataset bdd100k-scenario --checkpoint <weights> --data-dir <prepared dir> --split test.

Downloads last month
13
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for dronefreak/bdd100k-scenario-yolo26s-cls

Finetuned
(142)
this model

Dataset used to train dronefreak/bdd100k-scenario-yolo26s-cls

Collections including dronefreak/bdd100k-scenario-yolo26s-cls

Papers for dronefreak/bdd100k-scenario-yolo26s-cls

Evaluation results

  • Top-1 accuracy (test split) on BDD100K Scenario Classification
    BDD100K-Toolkit
    77.810
  • Macro F1 (test split) on BDD100K Scenario Classification
    BDD100K-Toolkit
    52.240
  • Macro precision (test split) on BDD100K Scenario Classification
    BDD100K-Toolkit
    58.280
  • Macro recall (test split) on BDD100K Scenario Classification
    BDD100K-Toolkit
    49.230