Datasets:
file_name stringclasses 5 values | quality stringclasses 4 values | sleeping_position stringclasses 5 values | age_estimation stringclasses 4 values | posture_points stringclasses 5 values | clothing_type stringclasses 5 values | blanket_presence stringclasses 4 values | head_position stringclasses 5 values | expression stringclasses 4 values |
|---|---|---|---|---|---|---|---|---|
0443ec3fa6e6f3d6804732a8ef008dfd.jpg | 1080*1368 | prone | about 3 years old | head, shoulders, hips, knees, feet | shorts and vest | none | right tilt | neutral |
72ee0ec436d27c353924b9d29af555d1.jpg | 1080*1374 | Side lying | 6-12 months | No coordinate information provided | Long-sleeve top, long pants | None | Left-turned | No expression |
9c9addf557a68dbdbada76bd900b96d8.jpg | 1080*1374 | side lying | 1-2 years | head, shoulder, hip, knee, foot | top and pants | none | turned left | no expression |
a1ed5fe28d7a7c9af13bcf65abf8ff1a.jpg | 1080*1392 | supine | approximately 3-5 years old | unable to provide specific coordinates | long sleeve top and pants | blanket present, but not covering | central | no expression |
f89657616846127b0a917bd425e86e7d.jpg | 1028*1438 | side sleeping | 1-2 years | no specific coordinates provided | onesie | no | right turned | no facial expression visible |
Infant Sleep Posture Recognition Image Dataset
Currently, there are many challenges in improving infant sleep safety, including the difficulty of assessing posture and monitoring sleep status without affecting normal infant sleep. Existing monitoring equipment often relies on clothing sensors, which may cause discomfort and excessive interference. This dataset aims to address the risk of asphyxiation caused by incorrect infant sleep posture through visual recognition technology, optimizing the data input structure of infant care monitoring systems. Images in this dataset were collected using high-definition camera equipment in a simulated home environment to ensure realistic and natural scenes. Quality control includes multiple rounds of annotation, consistency checks, and final review by childcare experts. The data annotation team consists of 30 professionals with backgrounds in medicine and artificial intelligence. Data preprocessing includes image standardization, data augmentation such as rotation and flipping, and is ultimately organized and stored in folders for easy access and application.
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| quality | string | Resolution |
| sleeping_position | string | The category of sleeping position identified, such as supine, lateral, or prone. |
| age_estimation | integer | Estimated age based on the physical features of the infant in the image. |
| posture_points | json | Coordinates of the keypoints of the infant's body posture identified in the image. |
| clothing_type | string | The type of clothing worn by the infant in the image, such as onesie or sleep sack. |
| blanket_presence | boolean | Indicates whether a blanket is covering the infant in the image. |
| head_position | string | Specific position of the infant’s head in relation to the body, such as center, left, or right. |
| expression | string | Facial expression of the infant in the image, such as smiling, crying, or neutral. |
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 |
Source & Contact
If you need more dataset details, please visit Mobiusi. or contact us via contact@mobiusi.com
- Downloads last month
- 5