3 ms·

New Camera Has Real-time Focus of Near and Far Field

2 points by J3L2404 17y ago

gjm11 17y ago
So, what they actually have is N cameras focused at different distances, plus a distance-measuring device; then the image they produce is basically a mosaic of the N cameras' images, choosing whichever one is nearest to being in focus. The clever bit is yet another camera, of a different kind, which somehow determines distance-to-camera at high resolution. Supposedly the key to this is the fact that intensity is proportional to 1/distance^2, but just how they use this (given that different objects will be different in brightness) the article doesn't say. Fortunately, the lead researcher has a press release at http://individual.utoronto.ca/iizuka/research/OmnifocusVideoCamera_newsrelease.pdf http://individual.utoronto.ca/iizuka/research/OmnifocusVideo... with a bit more detail: they illuminate the scene with two IR sources in turn, one closer than the other, and look at how brightness changes between the two. (So they have C/r^2 and C/(r+k)^2, where k is the known separation between the IR sources, r is the distance to the nearer source, and C is the IR reflectivity; from that they can compute r.) I guess this doesn't work so well with anything that doesn't reflect enough IR. At present N=2, and their demonstrations all show an image with some "near" and some "far" bits and nothing in between. Presumably they've carefully focused one camera on the "near" and one on the "far". Using N cameras in this way gets you N times the depth of field, and 1/N as much light into each camera. If instead you make your aperture N times smaller, you again get N times the depth of field, but now 1/N^2 as much light into the camera. So it does seem possible that it might be a win, for applications where you really want as much depth of field as possible, and nothing in your scene is moving much, and everything reflects enough IR.
- wendroid 17y ago
  > Supposedly the key to this is the fact that intensity is proportional to 1/distance^2, but just how they use this (given that different objects will be different in brightness) the article doesn't say. I'd imagine something not far away from : pixel(x, y) = max(intensity(x, y, ccds)) The IR distance system has been around for a while. Sony released a camera which recorded the distance per pixel too, they introduced it as an alternative to green screening, using the distance map as a matte. I can't find what that's called atm. and though I looked forward to using one, I've never come across one in my film work.
  - gjm11 17y ago
    No, not that (unless I'm misunderstanding you, which of course I might be). Here's what they do, according to the PDF that I linked to: 1. Illuminate with two IR sources, more or less on-axis and at known positions relative to the lens. 2. From the ratio of the resulting pixel intensities, estimate per-pixel distances. 3. Take each pixel from the image whose camera is focused nearest to that pixel's distance.
    - wendroid 17y ago
      Ah right. I realised in bed that what I guessed was totally wrong
wendroid 17y ago
No it doesn't. It has an array of CCDs and chooses the appropriate pixel to capture at runtime based on inferred distance information so the headline should read : "New Camera Simulates Real-time Focus of Near and Far Field" Sounds like a fun piece of kit though personally I think in general, having all objects in the field of view in focus will be sensory overload and perhaps a little disturbing, similar to the dissonance created by the inverse - fake depth of field.

2 points by J3L2404 17y ago

gjm11 17y ago

So, what they actually have is N cameras focused at different distances, plus a distance-measuring device; then the image they produce is basically a mosaic of the N cameras' images, choosing whichever one is nearest to being in focus. The clever bit is yet another camera, of a different kind, which somehow determines distance-to-camera at high resolution. Supposedly the key to this is the fact that intensity is proportional to 1/distance^2, but just how they use this (given that different objects will be different in brightness) the article doesn't say. Fortunately, the lead researcher has a press release at http://individual.utoronto.ca/iizuka/research/OmnifocusVideoCamera_newsrelease.pdf http://individual.utoronto.ca/iizuka/research/OmnifocusVideo... with a bit more detail: they illuminate the scene with two IR sources in turn, one closer than the other, and look at how brightness changes between the two. (So they have C/r^2 and C/(r+k)^2, where k is the known separation between the IR sources, r is the distance to the nearer source, and C is the IR reflectivity; from that they can compute r.) I guess this doesn't work so well with anything that doesn't reflect enough IR. At present N=2, and their demonstrations all show an image with some "near" and some "far" bits and nothing in between. Presumably they've carefully focused one camera on the "near" and one on the "far". Using N cameras in this way gets you N times the depth of field, and 1/N as much light into each camera. If instead you make your aperture N times smaller, you again get N times the depth of field, but now 1/N^2 as much light into the camera. So it does seem possible that it might be a win, for applications where you really want as much depth of field as possible, and nothing in your scene is moving much, and everything reflects enough IR.

wendroid 17y ago
> Supposedly the key to this is the fact that intensity is proportional to 1/distance^2, but just how they use this (given that different objects will be different in brightness) the article doesn't say. I'd imagine something not far away from : pixel(x, y) = max(intensity(x, y, ccds)) The IR distance system has been around for a while. Sony released a camera which recorded the distance per pixel too, they introduced it as an alternative to green screening, using the distance map as a matte. I can't find what that's called atm. and though I looked forward to using one, I've never come across one in my film work.
- gjm11 17y ago
  No, not that (unless I'm misunderstanding you, which of course I might be). Here's what they do, according to the PDF that I linked to: 1. Illuminate with two IR sources, more or less on-axis and at known positions relative to the lens. 2. From the ratio of the resulting pixel intensities, estimate per-pixel distances. 3. Take each pixel from the image whose camera is focused nearest to that pixel's distance.
  - wendroid 17y ago
    Ah right. I realised in bed that what I guessed was totally wrong

wendroid 17y ago

No it doesn't. It has an array of CCDs and chooses the appropriate pixel to capture at runtime based on inferred distance information so the headline should read : "New Camera Simulates Real-time Focus of Near and Far Field" Sounds like a fun piece of kit though personally I think in general, having all objects in the field of view in focus will be sensory overload and perhaps a little disturbing, similar to the dissonance created by the inverse - fake depth of field.