Traffic Lights of New York (TLoNY): Models

A comprehensive object detection dataset for fine-grained classification of traffic lights in New York City. The dataset includes both vehicular and pedestrian traffic signals with detailed color state annotations.

Overview

TLoNY provides accurately labeled traffic light data suitable for training object detection models, sourced from publicly available images from Unsplash. The dataset captures various lighting conditions, weather scenarios, and urban environments of New York City.

Metrics

Model Name Recall mAP50 mAP50-95
yolov8x-p2 0.50405 0.56544 0.30823
yolov8n 0.43263 0.50172 0.27866

Download

Dataset

Models

Cite

@misc {mehmet_kerem_turkcan_2025,
    author       = { {Mehmet Kerem Turkcan} },
    title        = { traffic-lights-of-new-york (Revision 2999230) },
    year         = 2025,
    url          = { https://huggingface.co/datasets/mehmetkeremturkcan/traffic-lights-of-new-york },
    doi          = { 10.57967/hf/4331 },
    publisher    = { Hugging Face }
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.