Why don’t cameras use time-of-flight depth sensing for autofocus?
Asked 6/18/2019
8 views
2 answers
0
Modern cameras use phase-detect and contrast-detect autofocus, but some devices can also measure distance with time-of-flight or other active depth-sensing methods. Why isn’t that commonly used for camera autofocus, and could depth data also be useful for creative editing or stitching? Are the main limits cost, power, range, or something else?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
7y ago
2 Answers
4
A simple answer:
Power.
A slightly less simple answer:
Power, and synchronization reliability.
Time-of-flight active range finding methods require you to project a signal out into the field, and to do so with something powerful enough to reflect back to the recording device with enough signal strength left to get useful readings off of.
That eats battery power, and while it may be useful in very low light or other challenging conditions, that is not a huge part of typical photography.
The other part is that knowing how far away something is from the camera doesn't actually do a lot for you when it comes to focusing a lens. In a carefully calibrated system it can be used in a round-about way to achieve focus, but it does not actually tell you anything about the focus.
So you can know that something is exactly 10.1543m away from the camera, and you can set the lens to the focus point that is was told should work for 10.1543m away, but this in and of itself doesn't tell you anything about whether or not the object is actually in focus... Since such a system does not rely on the focus of light through the image taking lens, then it has no means to confirm what the state of focus actually is.
Instead, camera manufactures have relied on systems that tie in closer to existing focusing methods. Canon has a 'focus aid' system built into many of their external flash units, which can project an IR grid into the scene to help the standard focus system lock on in low light for example. This grid is picked up by the standard optical focus system in the camera, and is based on how the lens is focusing light.
- Uses existing tech already in the camera.
- Uses power only on demand, assuming you remember to turn it off...
Originally by user82321. Source · Licensed CC BY-SA 4.0
user82321
7y ago
0
Generated from our catalog & community — verify before relying on it.
Time-of-flight depth sensing sounds attractive, but for most cameras its tradeoffs outweigh the benefits.
Main reasons:
- Power use: Active systems must emit a signal strong enough to bounce back reliably, which increases battery drain.
- Range limits: They’re most useful at short distances, so they’re less helpful with telephoto shooting or many general photography situations.
- AF usefulness: Knowing subject distance is not the same as knowing the image is perfectly in focus. It can help guide focus in a calibrated system, but it doesn’t directly measure focus accuracy the way on-sensor AF methods do.
- Complexity and reliability: These systems add hardware, calibration, and synchronization challenges.
- Cost and size: Extra sensors/emitters make cameras more expensive and potentially bulkier.
- Interference risk: Multiple active systems aimed at the same scene could interfere with one another.
As for extra RGB+depth creative uses, there is real potential, but products exploring richer depth capture have shown that technical promise alone doesn’t guarantee enough market demand to justify the added cost and support burden.
Recommended products
UniqueBot
AI7y ago
Your Answer
Related Questions
Why can phase-detect autofocus fail in low light even when high ISO still allows an exposure?
Why do high-end cameras still use some non-cross-type AF points?
Why can linear polarizers interfere with SLR phase-detect autofocus?
Do more phase-detect AF points improve focus accuracy on mirrorless cameras?
When do Nikon 1 cameras stop using on-sensor PDAF in low light?