but how do a lightfield captures light information in all of its space around it if its lense is relatively that small and how does it "reach" the frontal dimension to get the light information ahead of its position...? I still dont get it. edit : do they achieve that just by moving the camera forward ETC whilst recording with it?
A good talk, but I'm left wondering how light fields will relate to 3D realtime scenes, especially those with animation+vfx and how/if it can relate to object collision/interaction (i.e. will there need to be light fields and raycasting running simultaneously?)
Why is it called 4D when it is a 3D capture? Is the extra 3D information captured considered an additional dimension even though it is information pertaining to 3 dimensional space?
Actually I think this resembles our real world . Light is everywhere . Like in quantum Field . When light particles are excited we see the glow the color . It also has the momentum and is able to Generate matter movement , this the electricity and heat can be generated . We also live in the world written and created “spoken”. The matrix exist ))
Sounds like photography. I take a series of 2d photos from every angle and create a point cloud in the computer which then can be interpreted into a 3D mesh. The 2d images are also re calculated to plot out a texture map that will be wrapped over the geometry of the newly formed mesh of the object we made from our point cloud. I myself have run into the issues of specularity on a object throwing out my measurements of a subject.
Jonathan Doyle No, what you describe is a system that uses photographs to record the geometry of objects. A light field, as described here, uses photographs to record the geometry of light while completely ignoring the geometry of objects. A light field has no mesh just pixels. Consequently, if light field has a sufficiently high resolution it can reproduce correct specular highlights from every vantage point. If you understood light fields to be point-clouds then you fundamentally failed to understand what a light field is. Point-clouds and light fields are polar opposites.
3d reconstruction is not the same as lightfield rendering. Reconstruction has problems with reflectivity and specularity, whereas lightfield rendering captures and depicts it properly.
If I close my eyes I can still see a goat or anything I want to in my mind. My eyes are not actually seeing the object. If I dream I can see anything I want. But the images that we see in our dreams aren't real light - they're signals within our brains. So I want to ask what is imagination then? To "imagine" something is to use memory based on visual experience (accurate) and also based on no visual experience (inaccurate). If I never saw a goat before could I imagine it? Yes. Could I imagine it accurately? No. But if I am on the Star Trek holo-deck and I see a goat for the first time it will be correct because it is based on someone else's visual experience. In order to make hologram we need at least one creator because without a creator's own visual experience there cannot be a world. So if we are living an actual hologram this means we have a creator. What about someone blind from birth? Can they imagine? Blind people create a universe by touch and echolocation but this is not visual. However blind people's sensory data actually gets sent to and processed in the visual cortex. Sound information routes through the visual cortex. Their brains use echoes to generate spatial maps which for me is mind-blowing.
but how do a lightfield captures light information in all of its space around it if its lense is relatively that small and how does it "reach" the frontal dimension to get the light information ahead of its position...? I still dont get it.
edit : do they achieve that just by moving the camera forward ETC whilst recording with it?
How can we get in touch with this particular start up?
A good talk, but I'm left wondering how light fields will relate to 3D realtime scenes, especially those with animation+vfx and how/if it can relate to object collision/interaction (i.e. will there need to be light fields and raycasting running simultaneously?)
[ btw - I'm looking to use Light Fields within Unity when it's released... 'very exciting. :-) ]
May be even quantum computer can't process such amount of data in 3D space in high resolution...
Why is it called 4D when it is a 3D capture? Is the extra 3D information captured considered an additional dimension even though it is information pertaining to 3 dimensional space?
Array of cameras is 2D, each image captured by each camera is 2D, hence 4D.
Actually I think this resembles our real world .
Light is everywhere .
Like in quantum
Field .
When light particles are excited we see the glow the color . It also has the momentum and is able to
Generate matter movement , this the electricity and heat can be generated .
We also live in the world written and created “spoken”.
The matrix exist ))
Sounds like photography. I take a series of 2d photos from every angle and create a point cloud in the computer which then can be interpreted into a 3D mesh. The 2d images are also re calculated to plot out a texture map that will be wrapped over the geometry of the newly formed mesh of the object we made from our point cloud. I myself have run into the issues of specularity on a object throwing out my measurements of a subject.
Jonathan Doyle No, what you describe is a system that uses photographs to record the geometry of objects. A light field, as described here, uses photographs to record the geometry of light while completely ignoring the geometry of objects. A light field has no mesh just pixels.
Consequently, if light field has a sufficiently high resolution it can reproduce correct specular highlights from every vantage point.
If you understood light fields to be point-clouds then you fundamentally failed to understand what a light field is. Point-clouds and light fields are polar opposites.
3d reconstruction is not the same as lightfield rendering. Reconstruction has problems with reflectivity and specularity, whereas lightfield rendering captures and depicts it properly.
If I close my eyes I can still see a goat or anything I want to in my mind. My eyes are not actually seeing the object. If I dream I can see anything I want. But the images that we see in our dreams aren't real light - they're signals within our brains. So I want to ask what is imagination then? To "imagine" something is to use memory based on visual experience (accurate) and also based on no visual experience (inaccurate). If I never saw a goat before could I imagine it? Yes. Could I imagine it accurately? No. But if I am on the Star Trek holo-deck and I see a goat for the first time it will be correct because it is based on someone else's visual experience. In order to make hologram we need at least one creator because without a creator's own visual experience there cannot be a world. So if we are living an actual hologram this means we have a creator. What about someone blind from birth? Can they imagine? Blind people create a universe by touch and echolocation but this is not visual. However blind people's sensory data actually gets sent to and processed in the visual cortex. Sound information routes through the visual cortex. Their brains use echoes to generate spatial maps which for me is mind-blowing.
Replicant...
Geil Ziege in 3D. Genau mein Ding.
lytro got bankrupts so stop saying lytro lytro . lytro is dead .
very bad presenter
I thought he did a very nice job