Insights
Unlocking Metadata
Potential
Discover how detailed metadata enhances the value and the usability of our 3d human models.
Metadata
Understanding Our
Comprehensive Metadata
Collection
We have meticulously gathered metadata to provide detailed insights into each 3D model in our Human Scan Repository. This information enhances the usability and relevance of our datasets for various applications, including AI training, digital human modeling, animation, and research. Our metadata is categorized into two primary types: Personal Metadata and Pose Metadata.
Metadata
Metadata Types
We have meticulously gathered metadata to provide detailed insights into each 3D model in our Human Scan Repository. This information enhances the usability and relevance of our datasets for various applications, including AI training, digital human modeling, animation, and research. Our metadata is categorized into two primary types: Personal Metadata and Pose Metadata.
Personal Metadata
Our Personal Metadata ensures that each scanned individual is uniquely identified and classified based on key attributes. This helps in demographic studies, biometric applications, and realistic digital representation.
- Model ID – Unique ID identifier for each model
- AltID-01 – Alternative ID in format: Number[####]-Age[##]-Sex[#]-Ethnicity[#]
- Alias – Alias name for the person.
- Sex – Biological sex of the individual.
- Age – Age at the time of the scanning.
- Ethnicity – Four main ethnical categories.
- Country – The country of birth
- Roots – The ethnical country of origin.
- Height – The height of the individual in centimetres.
- Weight – The weight of the individual in kilograms.
- BMI – The Body-Mass-Index
- Body Type – Classification is based on the system by the World Health Organization.
- Eye Color – The eye colour represents a classification of the iris colour
- Hair Color – Categorized by checking the photos taken from the scan and doesn’t necessarily relate to the natural hair colour.
- Hair Type – Classification of the hair based on its shape and features
- Skin Type – The skin type is based on the Fitzpatrick’s Scale
- Features – This includes a list of any features detected on the person, such as tattoos, scars, piercing, beard, skin condition, varicose veins, stretch marks, and more.
Pose Metadata
Pose Metadata provides crucial details about each scanned pose, allowing users to understand how the model was captured, what they were wearing, and their possible intended action. This is particularly useful for gaming, fashion design, AI training, and digital human studies.
- Full Body Clothing – Outfits are being tracked that consist of several items that belong together, like uniforms, and would not appear individually in a regular outfit
- Upper Body Clothing – The upper body clothing item is tracked, like t-shirts or jackets.
- Lower Body Clothing – Similar to upper-body clothing, only clothing items worn on the lower body or legs that are visible are being tracked.
- Outfit Type – Describes the general look of the clothing the individual is wearing.
- Profession – Describes if the outfit is related to a corresponding profession.
- Season – Describes the season of the year this outfit is most likely being worn.
- Activity – Describe the action the individual might be performing in the capture
- Footwear – Describes the person’s footwear.
- Neck Accessories – Describes accessories worn on the neck and/or shoulders.
- Waist Accessories – Describes accessories worn around the waist, like belts.
- Wrist Accessories – Describes accessories worn on the wrist
- Hand Accessories – Describes accessories worn on the hands
- Head Accessories – Describes accessories worn either on the head or in the hair.
- Face Accessories – Objects such as piercings, glasses, or masks are noted for the face
- Expression – FACS-based expression coding AUXX is used for heads.
- Pose – The pose of the person at the capture.
- Primary Color & Secondary Color – This describes the one or two most prevalent colors on a single clothing item.
Context
Enhancing Results with Detailed Metadata
Enriched metadata is crucial to providing the right context to training AI models. Thats is why we have dedicated our time to providing accurate metadata for every 3D model, which will unlock your maximum training potential.
Metadata
Understand Our Comprehensive Metadatada Collection MEthodology for Enhanced Accuracy
We follow strict protocols and data collection standards to ensure that all metadata is precise, consistent, and useful for a variety of applications, from AI training to digital modeling.
Personal Metadata
To ensure accuracy and consistency, personal metadata was gathered through a structured process that combines direct measurement, standardized classification, and subject interviews. Models participated in a detailed intake session where key personal attributes were recorded. Height and weight measurements were taken using a calibrated digital scale. Hair, body, and skin types were classified using predefined classification systems, including the Fitzpatrick scale for skin tone, WHO standards for body type, and an internal categorization system for hair characteristics. These efforts ensure that the dataset maintains a high degree of accuracy and usability across various applications.
Pose Metadata
Pose metadata was collected through a meticulous review process combining automated image analysis and manual annotation by trained specialists. After the scanning process, images were analyzed to document clothing, accessories, and physical expressions. Each recorded attribute was verified by human annotators to maintain data integrity. Additionally, pose descriptions were classified using a standardized lexicon, ensuring consistency across the dataset. Manual annotation protocols were developed to minimize subjective bias, and routine quality checks were performed to refine metadata accuracy and completeness. This structured approach guarantees that pose metadata is both reliable and adaptable for a wide range of applications, including AI training, animation, and digital modeling.
Discover Our Enriched 3D Models
Explore our extensive datasets enriched with detailed metadata for your projects and research.