thundarstrom/traffic-surveillance-video-corpus

2K Dashcam Traffic Surveillance Video Corpus & CVAT Tracks

Part of the Edge-AI Traffic & Vehicle Analytics System repository by thundarstrom.

Dataset Summary

A continuous sequential video tracking dataset featuring 50 continuous 1-minute video drives in 2560x1440 (2K QHD) at 25 FPS (75,000 frames total).

Annotations Included

• Accompanied by 100 CVAT Video 1.1 XML annotation files (videoset1_xml).
• 514,471 tracked bounding boxes across continuous temporal IDs (rider, motorcycle, helmet, no_helmet, license_plate).
• Designed for training and evaluating DeepSORT / ByteTrack multi-object tracking, lane violation detection, and homography speed estimation.

---

How to Access and Download

Using Automated Project Downloader (Extracts automatically)

# Clone / pull and auto-extract dataset
python scripts/download_hf_datasets.py --dataset tracking_video --org thundarstrom

Using huggingface_hub Python SDK

from huggingface_hub import snapshot_download

# Download into local dataset directory
local_path = snapshot_download(
    repo_id="thundarstrom/traffic-surveillance-video-corpus",
    repo_type="dataset",
    local_dir="data/datasets/videoset1_videos"
)
print(f"Dataset downloaded to: {local_path}")

---

Recommended Training / Evaluation Recipe

python scripts/track_and_speed.py --video_dir data/datasets/videoset1_videos --xml_dir data/datasets/videoset1_xml

---

Citation & Maintainer

• Maintained by: thundarstrom
• Project: Edge-AI Real-Time Traffic Violation Detection & ANPR
• License: CC-BY-4.0