[Meshroom](https://github.com/alicevision/Meshroom) is the 3D Reconstruction Software built on AliceVision
### Industry Standard
### Industry Standard
...
@@ -104,22 +105,32 @@ iPhone and Android apps for photogrammetry and now LiDAR scanning have multiplie
...
@@ -104,22 +105,32 @@ iPhone and Android apps for photogrammetry and now LiDAR scanning have multiplie
# Light Field
# Light Field
<p>Light Fields are a new study field in photography. The objective is to capture the full plenoptic content of the scene, defined as the collection of light rays emitted from it, in any given direction. If the ideal plenoptic function was known, any novel viewpoint could be synthesized by placing a virtual camera in this space, and selecting the relevant light rays.</p>
Light fields represent an advanced form of passive sensing, aiming to capture full plenoptic content: all possible light rays emanating from a scene in any given direction. This results in a four-dimensional function, as it involves selecting a ray's position and angle. If the ideal plenoptic function was known, any novel viewpoint could be synthesized by placing a virtual camera in this space, and selecting the relevant light rays.
<p>In practice, we can only sample light rays in discrete locations. There are two popular optical architectures for this:</p>
<ul><li>Multi-camera systems: simply shoot the scene from several locations using an array of camera (or a single moving one).</li><li>Lenslets: a single CMOS sensor with an array of lenses in front.</li></ul>
In practice, we can only sample light rays in discrete locations. There are two popular optical architectures for this:
### Multi-Camera Systems:
Multi-camera systems: simply shoot the scene from several locations using an array of camera (or a single moving one).
### Lenslets:
Lenslets: a single CMOS sensor with an array of lenses in front.
<p>In the lenslet approach, each pixel behind a lenslet provides a unique light ray direction. The collection for all lenses is called a <strong>sub aperture image</strong>, and roughly corresponds to what a shifted camera would capture. The resolution of these images is simply the total number of lenslets, and the number of sub-aperture images available is given by the number of pixels behind a lenslet. For reference, the <ahref="https://en.wikipedia.org/wiki/Lytro">Lytro Illum</a> provides 15x15 sub-aperture images of 541x434 pixels each, which is a total of ~53 Megapixels.</p>
<p>In the lenslet approach, each pixel behind a lenslet provides a unique light ray direction. The collection for all lenses is called a <strong>sub aperture image</strong>, and roughly corresponds to what a shifted camera would capture. The resolution of these images is simply the total number of lenslets, and the number of sub-aperture images available is given by the number of pixels behind a lenslet. For reference, the <ahref="https://en.wikipedia.org/wiki/Lytro">Lytro Illum</a> provides 15x15 sub-aperture images of 541x434 pixels each, which is a total of ~53 Megapixels.</p>
<p><imgsrc="images/viewpoint7_png+img+margin.gif"alt="LF sub aperture images"></p>
<p><imgsrc="images/viewpoint7_png+img+margin.gif"alt="LF sub aperture images"></p>
<p>The most efficient layout for lenslets is hexagonal packing, as it wastes the fewest pixel area. Note that some pixels are not fully covered by the lenslet and receive erroneous or darker data. This means some sub aperture images cannot be recovered.</p>
The most efficient layout for lenslets is hexagonal packing, as it wastes the fewest pixel area. Note that some pixels are not fully covered by the lenslet and receive erroneous or darker data. This means some sub aperture images cannot be recovered.
<p><imgsrc="images/LF.png"alt="LF preview"></p>
<p><imgsrc="images/LF.png"alt="LF preview"></p>
<p>Light Fields have gotten a lot of traction recently thanks to their hight potential in VR applications. One impressive work was shown by Google in in a SIGGRAPH 2018 paper:</p>
Light Fields have gotten a lot of traction recently thanks to their hight potential in VR applications. One impressive work was shown by Google in in a SIGGRAPH 2018 paper:
https://www.youtube.com/embed/4uHo5tIiim8
<iframewidth="560"height="315"src="https://www.youtube.com/embed/4uHo5tIiim8"title="YouTube video player"frameborder="0"allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"allowfullscreen=""></iframe>
<p>Depth estimation on Light Field data is an active domain. For now, algorithms are commonly tested on ideal, synthetic light fields such as this <ahref="https://lightfield-analysis.uni-konstanz.de/">dataset</a>. Here is one example of point cloud obtained from a stereo<ahref="https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8478503">matching method</a>:</p>
Depth estimation on Light Field data is an active domain. For now, algorithms are commonly tested on ideal, synthetic light fields such as this [dataset](https://lightfield-analysis.uni-konstanz.de/). Here is one example of point cloud obtained from a stereo[matching method](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8478503).
<iframetitle="4D light field - depth estimation"frameborder="0"allowfullscreen=""mozallowfullscreen="true"webkitallowfullscreen="true"allow="fullscreen; autoplay; vr"xr-spatial-tracking=""execution-while-out-of-viewport=""execution-while-not-rendered=""web-share=""src="https://sketchfab.com/models/b9edfdd28c154ecf995da7b8c6590da8/embed"></iframe>
<iframetitle="4D light field - depth estimation"frameborder="0"allowfullscreen=""mozallowfullscreen="true"webkitallowfullscreen="true"allow="fullscreen; autoplay; vr"xr-spatial-tracking=""execution-while-out-of-viewport=""execution-while-not-rendered=""web-share=""src="https://sketchfab.com/models/b9edfdd28c154ecf995da7b8c6590da8/embed"></iframe>
