29 Sep 2023
Motion capture’s light-speed development has seen it branch out into more unexpected paths than anyone could have anticipated. Since its initial use for biomechanics research and clinical gait analysis in universities and hospitals, that same technology would soon go on to animate the world’s most memorable characters in film and gaming, revolutionize industrial practices, develop military hardware, and even help to build out virtual reality worlds including the metaverse.
The mocap world’s list of technical terminology has also grown exponentially. While it can be tough to keep up, it’s worth going back to basics to the most widely practiced format: optical motion capture. In this blog, we’ll delve into what optical motion capture means, and how it brings human movement to virtual life across a range of industries.
Motion capture is an example of photogrammetry, the practice of using photography for surveying purposes. In this case, cameras are used to measure small, bright dots of light within a whole capture space emitting from markers which are carefully attached to a person or object. Optical motion capture, which can also be referred to as ‘marker-based tracking’, uses a set of cameras to track the coordinates of these markers to construct a detailed three-dimensional view of a moving subject.
The majority of mocap systems use passive markers which ‘bounce’ light emitting from infrared LEDs circled around the cameras’ lenses, while other marker sets may use active LEDs, which instead give off their own light. The brightness of these markers ensure that they are the only images the cameras are able to pick up, rather than the test subject or any background “noise”.
Passive markers are usually retro-reflective and spherical, making it easier for a computer to work out their central points. When these central points are tracked by multiple cameras from different angles, they can be triangulated to produce 3D coordinates of the motion being performed. The resulting data can then be transposed onto a model or skeleton using mocap software.
Given the great level of detail gained by optical motion capture, using high-resolution cameras and involving minimal data cleanup, it is usually reserved for large-scale projects. It underpins the 3D animated characters featured in many big-budget films and TV shows such as Lord of the Rings, Avatar and Stargate SG1, as well as ‘Triple A’ computer games. These highly flexible systems can be used in large-scale indoor or outdoor spaces where a range of cameras can operate, such as a movie set or a laboratory. Biomechanics researchers, for example, can use optical motion capture to precisely measure the athletic movements of certain joints or muscles, or test the effectiveness of sports equipment.
Optical motion capture is a common method of marker-based tracking, and the quality of the capture is determined by the number of cameras. Some practitioners, however, use other mocap methods depending on their use case or project. Markerless systems, for instance, aim to complete the same task using software alone, rather than specialized tracking devices. However, they may not be as accurate as optical motion capture’s marker-based tracking for mapping high-resolution human movement.
Optical motion capture also differs from inertial motion capture, where subjects wear inertial measurement units (IMUs)––sensors strapped to their bodies or wearables that measure accelerations. Several stages translate movement into animation data. It’s a smaller setup acceptable for quick and easy motion capture but is limited in quality due to not measuring position directly.
Other methods include mechanical motion capture systems, which consist of an exo-skeleton structure attached to the test subject to approximate joint angles. Magnetic motion capture systems, less common nowadays, use sensors attached to the subject, which act as receivers to measure the low-frequency magnetic field generated by a transmitter. A computer then correlates the field strength within the capture space to calculate position, which is susceptible to errors caused by metal in the capture space.
While there is not a one-size-fits-all method, optical motion capture is an effective option for a range of use cases.
Recording motion data using optical motion capture requires multiple cameras for tracking purposes, a marker set, and processing software.
Cortex is our flagship motion capture processing software that uses optical systems for biomechanics, character animation, VFX, robotics, broadcasting, and more. Its compatibility with our Kestrel cameras allows for complex optical motion capture in large areas where robust equipment is needed for precise marker tracking. Alternatively, BaSixⓇ Go offers animators and other mocap artists a more affordable, lightweight optical motion capture option. Its range of accurate upgradable cameras are cross-compatible with various systems and work with active marker rigs.
Optical motion capture extends to every facet of movement analysis no matter the industry, letting filmmakers, visual artists, clinicians, sports coaches and more track motion, record mocap data, and construct valuable skeleton models for post-production and further research.
If you’re feeling inspired to find out more or explore our optical motion capture solutions, get in touch with our team today.