HDR motion shots

bs0604 · Nov 28, 2011

I just started using my Nikon d90 with Photomatix Pro to obtain/process HDR images. It works well but in view of having to take 3 pictures using the bracket feature on the camera, I don't understand how photographers take the action shots using HDR. eg I saw a picture of a horse rearing up that was in HDR but don't understand how this image was obtained in view of needing the many seconds to take a series of 3 bracket shots for Photomatix. Any help in educating me?

Rephargotohp · Nov 28, 2011

A lot of those are mearly "Tone Mapped " single images or they used a HDR Simulation program like Topaz Adjust or Lucis Effects.

However, I will say if you know how to use Photmatix's Manual De-ghosting tool it is possible to stop action on a subject, by deghosting a single frame out of the 3

Don’t tell me I can’t when I can

Bynx · Nov 28, 2011

That answer is about as good as you're going to get.

janok · Nov 28, 2011

You can process a HDR picture out of one single RAW-file in Photomatix. Then you avoid ghosting.

McNugget801 · Nov 28, 2011

janok said:
You can process a HDR picture out of one single RAW-file in Photomatix. Then you avoid ghosting.

Here we go again...

That's not True HDR, you're just tone mapping a single file.

janok · Nov 28, 2011

OK - I might be wrong.

Just out of curiosity - what is the definition of True HDR?

McNugget801 · Nov 28, 2011

janok said:
OK - I might be wrong. Just out of curiosity - what is the definition of True HDR?

Proper HDR is done with multiple exposures set at various spacing in order to capture a much wider dynamic range. If I take a single RAW shot then lighten the foreground 1/2 a stop is that considered HDR? Of course not, so why would tone mapping that same RAW shot be considered HDR?

Hope that makes sense.

Rephargotohp · Nov 28, 2011

janok said:
OK - I might be wrong.
Just out of curiosity - what is the definition of True HDR?

The whole idea is that we are trying to capture the dynamic rnage beyond what our Digital sensors can, We want to get closer to what our eyes can see and even beyond that. With that (dne with Multiple exposures) we can make a True High Dynamic Range Image (our 32bit file) Unfortunately, That files is not able to be displayed with any current Monitor or a Prints. So we must compress that data into what IS displayable and we do that with "Tone Mapping" .

Our hopes are that we make an image that fully fills the spectrum of a 16 bit Image. We are in the end not making a High Dynamic Range image but a full spectrum Standard Dynamic range image. The problem comes in with the way a lot of people Tone Map the image, that their images become mostly Midtones(check the histogram on Grunge style or Faux HDR images) and nothing more.

So their end result is actually a LOW dynamic Range image that is below what a standard Sensor is capable of. It may look more detailed (because of sharpening that are part of some algorithms and the fact that some shadows are now mid ranges so they add more edges).. But they are in fact Low dynamic range images

480sparky · Nov 28, 2011

janok said:
OK - I might be wrong.
Just out of curiosity - what is the definition of True HDR?

Technically, there's isn't one.

GRbenji · Nov 28, 2011

A HDR image, as I understand it, is a compression (aka tone mapping) of a high dynamic range scene (a HDR which the sensor can't capture in 1 exposure, hence multi exposure needed) into a DR which can be presented on print or monitor. Since the final product or output (known as HDR image) can be presented on print or monitor, it cannot be of high dynamic range anymore but what is important is it was derived from a HDR scene and shows details across the entire dynamic range.

EchoingWhisper · Nov 28, 2011

Most monitors can't display more than 9 stop dynamic range (I remember reading it somewhere), so basically any picture that exceeds 9 stops in dynamic range requires tone mapping and therefore is HDR. Although I don't tone map single images, HDR doesn't really means that you have to take several shots - high is relative.

janok · Nov 29, 2011

Interesting discussion

I have been doing HDR just for a year, and I’m far from an expert. I have only participated on a webinar with Trey Ratcliff, the rest of my experience comes from reading in forums like this - and try and fail (a lot of failing

As I understand – the RAW-file contains much more information than a “developed” JPG-picture. This means that it contains more information than it is possible to show on the screen as one single shot. If you like, you can develop 3 versions of a picture from the raw-converter, import them into CS5 and do the HDR-processing (automatically or manual). I believe this is the same principle Photomatix is doing when it creates a LDR picture out of a RAW-file.

I understand that you probably will get a better result shooting 3-5-9 pictures with different exposures, since there is a limit of how much information you can get out of one single RAW. But if one single RAW contains more information than it is possible to show on a screen – and you use a process that optimize the different parts of the picture to make a more realistic end-result, it seems like a true “HDR” to me. The result is usually not very compelling – since it usually creates more noise than using JPG.
Anyway – I guess this is a purely academic discussion. At the end it is the result that counts