We propose a visual marker embedding method for the pose estimation of a projection surface to correctly map projected images onto the surface. Assuming that the surface is fabricated by a full-color or multi-material three-dimensional (3D) printer, we propose to automatically embed visual markers on the surface with mechanical accuracy. The appearance of the marker is designed such that the marker is detected by infrared cameras even when printed on a non-planar surface while its appearance can be diminished by the projection to be as imperceptible as possible to human observers. The marker placement is optimized using a genetic algorithm to maximize the number of valid viewpoints from which the pose of the object can be estimated correctly using a stereo camera system. We also propose a radiometric compensation technique to quickly diminish the marker appearance. Experimental results confirm that the pose of projection objects are correctly estimated while the appearance of the markers was diminished to an imperceptible level. At the same time, we confirmed the limitations of the current method; only one object can be handled, and pose estimation is not performed at interactive frame rates. Finally, we demonstrate the proposed technique to show that it works successfully for various surface shapes and target textures.

Electrically tunable lenses (ETL), also known as liquid lenses, can be focused at various distances by changing the electric signal applied on the lens. ETLs require no mechanical structures, and therefore, provide a more compact and inexpensive focus control than conventional computerized translation stages. They have been exploited in a wide range of imaging and display systems and enabled novel applications for the last several years. However, the optical fluid in the ETL is rippled after the actuation, which physically limits the response time and significantly hampers the applicability range. To alleviate this problem, we apply a sparse optimization framework that optimizes the temporal pattern of the electrical signal input to the ETL. In verification experiments, the proposed method accelerated the convergence of the focal length to the target patterns. In particular, it converged the optical power to the target at twice the speed of the simply determined input signal, and increased the quality of the captured image during multi-focal imaging.

Projection-based tabletop sharing (PBTS) systems allow a local user to share a pointing gesture or a handwritten note on a tabletop document with a remote user who places the same document on a tabletop by projecting the upper limb image of the local user onto the remote document and tabletop. A vertical display is used to share an image of the upper body, including the face of the remote user. However, in previous systems, the spatial layouts of the shared documents must be identical on both tabletops, and the projected upper limb is not extended from the upper body image of the remote user. This paper proposes a PBTS system to address such geometric consistency issues to improve remote collaboration. First, we propose to maintain the geometric consistency of a pointing gesture and handwritten note between each pair of the shared documents rather than between the entire tabletops. This allows users to freely change the document layouts on both tabletops. Second, we propose to overlay the upper limb image such that it is extended from the vertical display, where the upper body image is shown. This is achieved by rotating the upper limb image around the fingertip that performs the pointing gestures or around the tip of the pen used to write notes. We constructed a prototype to determine if the proposed system resolves the geometric consistency issues. Then, we evaluated how accurately a user can convey a pointing position to a distant partner when the document layouts differ between the remote tabletops. Finally, we evaluated how the user experience, particularly the social presence, is improved by the proposed geometrically consistent upper limb direction.