Taking away luck in street photography ?

[NY_Dan]

This is everything to do with the technology and the quality of single frames pulled from video, but if that quality was indistinguishable from a stills camera how would you know how it was shot.

Well, I believe one of the OP's points is that high res video can make almost anyone capable of producing great street photos -- the premise being that with the camera constantly on something good will be captured, or at the very least, the photographer will not have to choose a perfect/decisive moment to shoot.

Now some have opined that the camera won't be on continuously but only for 2-3 second bursts -- well then, a motor drive still camera would achieve the same result. To use a 4K camera as a still camera is possible, but then why not just use a still with a motor drive. I agree, one can't always tell what recoding medium was used to make an image.

 

[bobbyrab]

Well, I believe one of the OP's points is that high res video can make almost anyone capable of producing great street photos -- the premise being that with the camera constantly on something good will be captured, or at the very least, the photographer will not have to choose a perfect/decisive moment to shoot.

Now some have opined that the camera won't be on continuously but only for 2-3 second bursts -- well then, a motor drive still camera would achieve the same result. To use a 4K camera as a still camera is possible, but then why not just use a still with a motor drive. I agree, one can't always tell what recoding medium was used to make an image.

Actually what he said was you could shoot for 40 mins and get 72000 frames, but he then suggested that it could be used by a street photographer using it more discriminately, which I've then hijacked as short video bursts.
I was thinking of how this would play out once the technology improves, who knows where it will be in 20 years, as for now, no, it's not there yet, but if the frame rate could be improved sufficiently it would certainly be quieter than a motor drive.

 

[Darthfeeble]

Spray and Pray taken to the extreme. I like technology and have embraced digital totally but this isn't photography, it'd videography. I have a hard time accepting "the decisive moment" when plucked from a video feed.

 

[craygc]

You are essentially always shooting at 1/24 (for cameras that shoot 24 frames per second). Not enough to freeze fast moving subjects. Though, of course, not every photo needs this.

This is not correct. Video does not shoot continuously, it shoots discreetly. At 24 frames per second, each frame might, for example be 1/125 each. This would mean that you would have 1/125 of image capture time per frame followed by 4/125 of no capture time between frames. This is why in my previous post I was highlighting the ineffectiveness of this approach - @1/125 you are only capturing images 20% of the time.

 

[Darthfeeble]

Oh, and it isn't luck in the hands of a good street photographer.

 

[v_roma]

Thanks, I stand corrected.

This is not correct. Video does not shoot continuously, it shoots discreetly. At 24 frames per second, each frame might, for example be 1/125 each. This would mean that you would have 1/125 of image capture time per frame followed by 4/125 of no capture time between frames. This is why in my previous post I was highlighting the ineffectiveness of this approach - @1/125 you are only capturing images 20% of the time.

 

[katrak]

I read these comments with fascination !

To be clear, anyone who presses record, goes for a walk for an hour and expects a a dozen (or even one) masterpiece(s) awaiting him or her is utterly deluded.

While this technology will eventually make some impact on the street photography genre it will never eradicate the cold hard slog of looking for the elements of composition, sizing up a vantage point, waiting for the actors to appear on stage or for the light to be just right

But capturing fleeing gestures, glances and movements is different. Two people gesticulating wildly on a street corner ? A man running after a dog running after a rat ?

Used with restraint this technology will make a difference (and especially for documentary purposes) and in 5 years time, you could probably extrapolate the current 8mp jpegs to 24mp Raws.

Some technical points you've brought up - although the frame rate is 25 or 30 per second, you can change the duration of each frame's exposure to up to 1/6000 of a second - so in effect you end up with 30 pictures frozen at that speed.

And yes, you could leave the camera running like this until the SD card is full (64GB equates to 40 minutes).

As for editing, using Lightroom you can play the video at normal speed and when something grabs your attention freeze it, jiggle the frames forward and back and then grab the one you want. In this way 72,000 theoretical frames takes about an hour or so to cull.

Don't forget that some of the great street photographer took frame after frame to the point of obsession - Winogrand took 5,000,000 plus pictures in his life time (and died not having looked at 400,000 of them)

But it's early days....I've only been out for 30 minutes with this toy so far...

 

[Sparrow]

... but then sometimes street photographers strike back ... and still looked dead cool as they did it

 

[mickbenjamins]

This is not correct. Video does not shoot continuously, it shoots discreetly. At 24 frames per second, each frame might, for example be 1/125 each. This would mean that you would have 1/125 of image capture time per frame followed by 4/125 of no capture time between frames. This is why in my previous post I was highlighting the ineffectiveness of this approach - @1/125 you are only capturing images 20% of the time.

Good info, thanks for pointimg this out 🙂

 

[JoeV]

@NY Dan: With the Lumix G series cameras a person doesn't have to put the camera to one's eye to compose. You can flip out the rear LCD for waist-level viewing. Like Vivian Meier at 24 shots per second!

If I were doing this, I'd only record video when a likely scene was coming together. It would more resemble one of Winogrand's contact sheets.

Of course, someone will likely make an app that uses scene recognition to determine the "best" street composition out of the thousands of likely candidates. Maybe a "Garry Mode," or "Henri Mode," or "Daido Mode" to choose from!

~Joe

 

[thegman]

I think technically, if we assume that street photographers capture moments, without interfering too much, then you could get excellent photos just by leaving a few HD cameras on record in strategic places around a city. I can't really think of any reason why that would not be true, perhaps lovers of razor thin DOF might not get what they're after, but other than that, why could it not work?

It does sound like a very tedious hobby though.

 

[Richard G]

Interesting. Will likely have it's uses, like burst mode. Overgaard says he has his M9 shutter release always set to C, continuous. One problem with the video frame approach is the opportunity cost of being pointed in one direction waiting for the decisive momemnt, rather than having the camera inactive, using one's eyes, with FOV of >180˚, ears and nose and other senses to sniff out the next photograph.

 

[giellaleafapmu]

I think I will try this technology when it will be a bit more mature for sport. There I can see
a clear advantage since we have only a limited reaction time (going down each year) and athletes and their devices are getting faster and faster every year...

For casual street photography it sounds like a torture to me...and not that fun after all.

GLF

 

[jsrockit]

I have a hard time accepting "the decisive moment" when plucked from a video feed.

Was the real purpose of "the decisive moment" to show your prowess at target practice? or was it to make a great photograph using the technology of that time?

 

[mfunnell]

I find it painful enough going through the "take" from the day using a still camera. (Digital vs film hasn't made that much difference either: I can recall going through 2-4 rolls of 36 in a day, frequently, when it was much easier to have C-41 developed and scanned in same-day processing.) If going through 140-odd frames is/was scary, the thought of trying the same thing with 387,491 stills captured from a video feed makes various of my parts retract way too far.

...Mike

 

[williams473]

A skilled photographer could absolutely produce good images using this method, but the end product is still affected by choice: where the video camera was pointed, lighting, framing, timing, operator skill (exposed well, under/over etc.) - all still apply. And then once this massive pile of visual information is processed, the editing would be the next crucial step determining the nature of the final image. This is of course not a new concept - but taking stills from motion pictures works if the motion pictures are well conceived.

 

[taxi38]

Why not use somebody elses video feed,......stay in and watch the football.

 

[Roger Hicks]

Why not use somebody elses video feed,......stay in and watch the football.

The perfect analysis! Unless you know when to press the button (having THOUGHT beforehand), it doesn't matter whether you take 1 photo/second or 100 photos/second.

Cheers,

R.

 

[Carterofmars]

Did Ansel roam Yosemite with a super 8?

 

[DougFord]

Of course, someone will likely make an app that uses scene recognition to determine the "best" street composition out of the thousands of likely candidates. Maybe a "Garry Mode," or "Henri Mode," or "Daido Mode" to choose from!

~Joe

Agreed. Any idiot can leave their google glasses in record mode all day. The software will determine a 'hit or a miss'. This sort of recognition software could even contain a biomarker of the designer or someone no longer with us. Maybe we could dig up Hank? We're gonna need a dna sample. (lol)