News and Thoughts Regarding a One Lens 3D Camera
The news referred to in the article title is that ISee3D has announced the availability of single lens 3D camera technology. Now, let’s back up a bit.
One way to create a single lens 3D camera is to position a mechanism between the lens and the imager that can sequentially occlude half of the lens aperture. Doing so has the effect of "moving the center" of the lens. By shifting the center, the image path is divided into a left and a right perspective and allows the camera to capture a stereo pair.
Single lens 3D technology has been around for some time. The ISee3D web site refers the company’s ownership of a basic patent on the technology. It is US 5,828,487 which has a filed-for date of June 6, 1996. The technology was first applied to 3D endoscopes where the mechanism was a vibrating mechanical shutter. In the just announced ISee3D embodiment, the mechanism is a liquid crystal optical shutter.
The approach should produce well-matched image pairs because both images are transmitted through one lens, although there are slight differences in the light path on each side of the lens. Vertical, horizontal and rotational alignment are inherently accurate. Focus and zoom should be as easy to accomplish in the 3D mode as in the conventional 2D mode.
More than that, the production of well-matched image pairs eliminates the need for post capture software processing. This is especially important in real time applications where minimizing camera set up related artifacts should also minimize the occurrence of associated negative physiological effects in viewers.
It might also be added that the one lens 3D approach has the potential of being simpler and therefore less expensive that a conventional 2 lens approach.
A final positive was explained by Dwight Romanica the CEO at ISee3D: the technology can scale across device sizes. This means that the technology can be used in cell phones, digital cameras and camcorders.
On the down side, the liquid crystal shutter likely has a transmission of about 30%. Add this to the fact that the aperture of the camera’s optical system is at least halved for each of the stereo images. The result is a heavy hit to the light level presented to the imager. The light loss is somewhat mitigated by the fact that sensor capability has increased dramatically in recent years.
Another issue associated with the approach is that the parallax from a "reasonably" sized lens is small, which means the camera is best suited for imaging small objects that are close to the camera.
ISee3D’s business model is to license technology. The company seeks partners to jointly develop products incorporating single lens 3D technology. In fact, ISee3D has just announced achieving its first revenue with the billing of engineering fees for the development of a prototype for an unnamed camcorder manufacturer. And, we were told in a previous meeting, that CES should be a good place to see cameras based on the technology.
So, how does this all add up? Does one lens 3D have the potential to be a winner in commercial products? In my opinion, the answer can be yes but... to be a winner, the approach must enable products at least as good and costing no more than equivalents based on more conventional 3D approaches. Examples of the competition include cameras with two lenses in one body such as the Fuji FinePix and the new Panasonic camcorder.
More information:
Description of ISee3D Stereoscopic Vision System Natural 3D
By Art Berman, Display Daily
