Note

This is a Hugging Face dataset. Learn how to load datasets from the Hub in the Hugging Face integration docs.

Hugging Face

Dataset Card for ThermalPersonDetector#

A thermal image dataset for detecting people in a scene. The dataset contains only one class person

This is a FiftyOne dataset with 8778 samples.

Installation#

If you haven’t already, install FiftyOne:

pip install -U fiftyone

Usage#

import fiftyone as fo
import fiftyone.utils.huggingface as fouh

# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = fouh.load_from_hub("dgural/Thermal-Person-Detector")

# Launch the App
session = fo.launch_app(dataset)

Dataset Details#

Dataset Description#

Here are a few use cases for this project:

Sports Analytics: The “PersonDetection” model could be used to analyze individual athletes’ performances in various sports such as skateboarding, basketball, or soccer, by detecting and tracking the movements of players.

Surveillance Security: It could be utilized in CCTV systems and security cameras. By recognizing people in real-time, it could alert security personnel when unauthorized individuals are detected in restricted areas.

Social Distancing Detection: In light of the Covid-19 pandemic, this model could be used to enforce social distancing measures by tracking the number of people and their relative distances in public spaces like parks, malls, or transportation systems.

Smart Home Management: It can be deployed in smart homes devices to recognize the home occupants and subsequently adapt the environment to their preferences such as lighting, temperature or even play their favorite music upon entry.

Motion Capture and Gaming: In the gaming and animation industry, this model could be used for real-time motion capture, allowing developers to create more realistic and immersive human characters.

  • Language(s) (NLP): en

  • License: CC 4.0

Dataset Structure#

One ground_truth field with a person class

[More Information Needed]

Source Data#

Roboflow Univers

Who are the source data producers?#

SMART2

[More Information Needed]

Who are the annotators?#

SMART2

BibTeX:

@misc{
persondection-61bc2_dataset,
title = { PersonDection Dataset },
type = { Open Source Dataset },
author = { SMART2 },
howpublished = { \url{ https://universe.roboflow.com/smart2/persondection-61bc2 } },
url = { https://universe.roboflow.com/smart2/persondection-61bc2 },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2023 },
month = { may },
note = { visited on 2024-07-19 },
}