The lightfield cameras redefining images

In a world of greater immersion and picture-perfect realism, traditional high-resolution 2D image and video could be rendered all but obsolete in the near future. Javier Elizalde explains how Wooptix’s new single lens light field camera and development kit could redefine the very concept of what an image is.

The world of displays and imaging is one of constant change and evolution. In the last 10 years, the jump from1080p to 4K and onto 8K and beyond has occurred at breakneck speed.

While the display technologies continue to soar to new heights, the future of camera imagery itself is evolving in eye-opening new ways.

Wooptix, a developer of lightfield and wave front phase imaging platforms is on a journey to redefine what an image is, steering away from the conventional, high-resolution 2D image on a screen to a fully fleshed out 3D image with depth and colour correction.

What’s the secret? Light field camera technology.

Whereas a traditional camera only captures light intensity, light field cameras capture information about the light field present in a scene, capturing the direction of the light rays as they travel in space.

However, the team at Wooptix feel that the existing light field technologies on the market do not live up to the full potential of the technology.

Light field cameras are, currently, not without their limitations. A constant battle between full resolution and other areas means trade-offs need to be made as Elizalde clarified: “A full-resolution, single lens lightfield camera comes at a cost. If you get the full resolution, you end up with limitations in the other fields that conventional light field cameras have. Today’s limitation is the field of view. With a lightfield camera, you can capture an image or video and process it later. Then you can work on focus, changing point of view, fitting 3D or light field displays, selecting a plane, inserting digital objects in the plane and more.

“The problem with conventional light field technology is the resolution because you are putting microlenses in front of the sensor. You divide the resolution by the number of micro lenses that you have. To have a 2K image, you need a 16K camera to record a 2K image.

“Conventional light field cameras use a technique which has a lot of micro-lenses in front of the sensor. We come from astrophysics and astronomy where we have been playing with micro-lenses for a long time by using light field technology and looking at the stars to acquire the light rays’ direction. The problem with micro lenses is the resolution. When you have low resolution, you cannot play with the stars. We inverted the technique, as when you are acquiring a light field, you are acquiring the image in focus and the depth map. We asked: ‘If the result we need is a depth map and in focus image, why don’t we focus that image and process it in real time?’ As we know the distances, we are able to reconstruct the depth map so that we have a depth map and image in high resolution.”

""In the future, you are going to be able to feed any display with all the information in a scene." - Javier Elizalde, Wooptix. 

Enter Wooptix’s SEBI camera and development kit, a high-resolution light field camera that can obtain depth and colour information in real time by extracting information in real time from the focal stack of every pixel in the image sensor. The camera can acquire 156 frames per second and capture depth information at six different focal distances to generate images and depth maps at 26 frames per second.

Instead of using a conventional micro lens array, Wooptix’s SEBI uses a liquid lens in the optical chain in front of the sensor, allowing for the rapid changing of the focal plane that the sensor is focused on to capture multiple image planes at 30 frames per second.

Elizalde said: “The camera that we released is a development kit for early adopters. We are targeting companies that develop hardware for XR, virtual effects for Hollywood movies and videoconferencing because companies could be controlling a videoconference and providing more reality to conferences. The field of view in conferencing means that the camera only needs to capture the head and the shoulders, which is perfect for this technology.

“Our main target is currently video conferencing. We have a liquid lens in front of the camera, but the image is not processed within the camera but is instead processed in a connected GPU. The GPU processes the image as the amount of information inside the image is quite large.

Elizalde explained: “With this development kit, we can do videoconferencing, telemedicine, endoscopy (because of the small lens) and special effects to a certain degree with the limitation of field of view. If you want to do short things like animation, then it can also work very well.

Wooptix is currently expanding its horizons and aiming to generate further investment to develop the technology further, targeting larger companies and overcoming hurdles to bring the technology to a broader market.

Elizalde: “In the past four years, we have received investment from Intel Capital who have been reaching out to big companies. We developed the technology and needed to prove to Silicon Valley that the technology was good, especially for a Spanish company, a country that is not as well known for its technology as it is for its tourism industry. We have raised up to €7 million and we’re aiming to raise around 20 million next year.”

Looking to the future, Elizalde identifies light field cameras (otherwise known as volumetric or 3D cameras) as the everyday camera of the future, with Wooptix setting its sights on new horizons in the coming months.

Elizalde: “In the next three months, we will be working on a mobile form camera with a bigger field of view so we can develop bigger cameras that also work in real time. We also plan to work with two or three cameras at the same time, getting the light field of every camera to get more information from three lightfield cameras than you would from 14 2D cameras.

“In the future, you are going to be able to feed any display with all the information in the scene. In the future, people are going to use 3D capture, and in five years everything will be about volumetric/light field cameras. People are going to be using 3D/light field captures commonly in the future and I think that, in five years’ time, we’re going to see a lot of these cameras.

“We need companies to test our technologies and give us feedback to generate the next generation of video cameras, that is our ultimate goal.”

 

Article Categories