Visual Hull is a technique in computer imaging that creates a 3D shape of an object from multiple 2D images taken at different angles. Silhouettes are traced from the images and stitched together to form a 3D object. The process is less expensive and processor-intensive than some other techniques and can be used for obstacle detection, motion capture, and scanning virtual 3D objects. The accuracy of the technique depends on the sophistication of the algorithms and the controlled conditions under which the images are taken.
In computer imaging, a visual shell is the three-dimensional (3D) shape of an object that is extrapolated from multiple two-dimensional (2D) images taken at different angles around the approximated object. Surface data about an object’s shape is obtained by tracing the outline of an object in an image, essentially creating a silhouette of the object with no specific internal texture or detail. A collection of silhouettes, all extracted from images taken from different angles, are stitched together in 3D space, and the area between known contour points is interpolated to form a 3D object that has the general 3D contour of the real object, even though perhaps without so much specific detail. The process used to create a visual hull, also known as shape-from-silhouette (SFS), can be faster, less processor intensive, and less expensive to implement than some stereoscopic techniques for capturing 3D motion or shape sensing of 3D objects. Some of the applications that use Visual Hull include obstacle detection using computer vision, motion capture for medical or analytical purposes, and scanning virtual 3D objects when SFS is performed under highly controlled conditions.
The process of forming the visual envelope of an object from a set of images involves isolating the outline of the object from the background in the images. The exact location and orientation of the cameras used to capture the images is also important to the process. In each image, a straight path is drawn from the image viewing plane to the scene space and ends at the boundaries of the imaged object. This is done for each image and the area where each of the paths, which look like cones in a 3D environment, cross gives a very rough, block-like volume that contains the object within the dimensions of the scene . For some applications, such as computer vision, this information is sufficient for basic obstacle avoidance.
Silhouettes can be further refined so that the smallest geometric details are translated into the visual hull. These can include holes in the object, as might happen if the visual hull were constructed from images of a human standing with legs apart or arms outstretched. One attribute of an object’s shape that cannot be accurately captured with SFS techniques is a concave surface, because it does not contribute to silhouette.
The SFS technique for creating the visual envelope of an object can be incredibly detailed and accurate if sophisticated algorithms are used in combination with controlled conditions to create the source images. These conditions can include a single coherent light source, a static, measurable background, and precisely calibrated cameras. Given these conditions, very accurate 3D models of objects can be constructed, and motion capture can be performed without the need for markers, tracers, or special equipment other than cameras.
Protect your devices with Threat Protection by NordVPN
