Living Room Smart Home Abnormal Behavior Detection Dataset

#Behavior Recognition #Anomaly Detection #Video Analysis #Smart Home #Security Monitoring #Anomaly Detection

500 records
1.2G
MP4
CC-BY-NC-SA 4.0
MOBIUSI INC

Updated:2026-03-14

AI Analysis & Value Prop

With the popularity of smart home devices, accurately detecting abnormal behavior in the living room has become a major focus for the industry. Current solutions often rely on traditional video surveillance; however, their accuracy and real-time capabilities are limited and susceptible to changes in lighting and device malfunctions. This dataset aims to improve the perception of abnormal behavior by smart devices in real household environments, meeting business needs for security monitoring and user behavior analysis. Data is collected through cameras arranged in home living rooms, with the environment set to common daily layouts. Quality control measures include multiple rounds of annotation and consistency review by a team of over 20 experts in computer vision and behavior analysis. Data preprocessing includes steps such as denoising, keyframe extraction, and behavior feature extraction, with all videos ultimately organized and stored by date and event type in MP4 format.

Dataset Insights

Sample Examples

e909ae67**.mp4|720*1280|814.96 KB

9ec5bdc7**.mp4|720*1280|319.33 KB

Technical Specifications

Field	Type	Description
file_name	string	File name
duration	string	Duration
quality	string	Resolution
activity_type	string	The type of activity detected in the video, such as walking, falling, etc.
activity_duration	float	The duration of a specific activity occurring in the video, measured in seconds.
person_count	integer	The number of people appearing simultaneously in the video.
object_interaction	string	The interaction between a person and objects, such as opening a door, picking up an item, etc.
lighting_conditions	string	The lighting conditions during the video recording, such as bright, dim.
sound_level	integer	The average sound level in the video, measured in decibels.
camera_angle	string	The description of the camera angle during shooting, such as overhead, eye-level.
emotion_detection	string	The emotional state detected in individuals, such as happy, angry.

Compliance Statement

Authorization Type	CC-BY-NC-SA 4.0 (Attribution–NonCommercial–ShareAlike)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What is the Living Room Smart Home Anomaly Detection Dataset?: The Living Room Smart Home Anomaly Detection Dataset is a video dataset used to enhance the anomaly detection capabilities of smart devices.

What is the main purpose of this dataset?: The main purpose of this dataset is for anomaly behavior detection in smart devices, enhancing their ability to recognize abnormal situations.

What types of data does the dataset include?: The dataset includes video data showcasing potential anomalous behaviors in a living room environment.

Why do smart home devices need to use an anomaly detection dataset?: By using such a dataset, smart home devices can better learn and identify anomalous behaviors, thereby enhancing their security and intelligent service capabilities.

What are the advantages of the Living Room Smart Home Anomaly Detection Dataset?: This dataset provides a wealth of videos on anomalous behaviors in living room scenarios, helping smart devices to perform anomaly detection more accurately.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Living Room Smart Home Abnormal Behavior Detection Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/416b6eeb605e1e1dae6e8d8012e657fb?dataset_scene_id=6},
  urldate={2026-02-04},
  keywords={Smart home anomaly detection, video behavior recognition, smart device video dataset},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Products

Scene

Domain

Modality

Task