biometrics identification from body shape

biometrics identification from body shape

The science of identifying people from their physical characteristics is called biometrics. Biometrics is becoming a relevant field in science for several reasons, the least of which is the heightened demand for security in various situations. Biometrics use in computer science as a form of identification and access control. Biometric identifiers are the distinctive, measurable characteristics used to label and describe individuals. Physical biometrics, such as fingerprints, facial recognition, and Iris scans, are currently used for security purposes. However, so-called behavioral biometrics, such as gait recognition. Anthropometry plays an essential role in industrial design, clothing design, and architecture, where statistical data about the distribution of body dimensions are used to optimize products. Human recognition refers to the measurement of the human individual. Anthropometry involves the systematic analysis of the physical properties of the human body, primarily dimensional descriptors of body size and shape.

Figure 1: Landmarks on body
Figure 1 shows landmarks on the naked body under clothes. Red vertices are used from the front view, while yellow ones are used for the side view.

CAESAR: Civilian American and European Surface Anthropometry Resource

CAESAR began in 1997 as a partnership between government and industry to collect and organize the most extensive sampling of consumer body measurements for comparison. The project collected and organized data on 2,400 U.S. & Canadian and 2,000 European civilians, and a database was developed. This database records the anthropometric variability of men and women, aged 18–65, of various weights, ethnic groups, gender, geographic regions, and socioeconomic status.

The study was conducted from April 1998 to early 2000 and included three scans per person in a standing pose, full-coverage pose, and relaxed seating pose. Data collection methods were standardized and documented so that the database can be consistently expanded and updated. High-resolution measurements of body surfaces were made using 3D Surface Anthropometry. This technology can capture hundreds of thousands of points in three dimensions on the human body surface in a few seconds.

Gait recognition

For the first time, biometric gait recognition was proven as Visual Motion Perception by the American scientist Gunnar Johansson in 1975. Gait recognition is to identify humans based on their gait features. Compared with other biometrics such as the face, iris, palm print and fingerprint, gait features are still obtainable and recognizable at a distance with a low-resolution video.

Given many scans of a single person, we align them all with a unique, individualized model using a process we call co-registration. Co-registration simultaneously aligns all the person's scans to the model while building a statistical model of the person used, in turn, to constrain the registration. This process results in a digital avatar that captures the detailed shape and poses a variation of the subject. This avatar can be reposed and animated with realistic soft-tissue deformations.

Three-dimensional (3D) shape models are powerful because they enable the inference of object shape from incomplete, noisy, or ambiguous 2D or 3D data. For example, realistic parameterized 3D human body models have been used to infer the form and pose of people from images. The first step in training such models involves registering a database of 3D scans with a standard 3D template. Once all the scans are in alignment, one can build powerful statistical models of shape.

It has many advantages over the old measurement system using tape measures and other similar instruments. It provides details about the surface shape and 3D locations of measurements relative to each other and enables smooth transfer to Computer-Aided Design (CAD) or Manufacturing (CAM) tools.

Watrix - Biometric Gait Recognition
Figure 2: Watrix - Biometric Gait Recognition

Watrix is so far the only commercialized Gait Recognition Company in the world. It developed the world's most advanced gait recognition technology. The average recognition rate can reach 94.1%, much better than the previous best result (less than 65%). Watrix also built the world's largest gait database. Moreover, the company took a leading position in all aspects of this technology, such as target detection, segmentation, recognition, and tracking. In this way, it can recognize people from 50 meters away using ordinary 2K cameras and have less correlation with face information and lighting conditions.

Usage of the 3D human body Shape:

Gait recognition system used for airport security. The way you walk and your footsteps could be used as a biometric recognition at airport security instead of fingerprinting and face. Scientists working for private companies and government agencies conduct anthropometric studies to determine a range of sizes for clothing and other items. Measurements of the foot are used in the manufacture and sale of footwear. Other areas from the usage of 3D human body shape include:

  • Graphics: animation for games and films.
  • Fitness: how diet and exercise effects shape.
  • Ergonomics: how bodies fit products?

Related posts: