0
\$\begingroup\$

I'm not new to photography (I have been doing it for some years) but I have been doing some HDR for the last couple of months.

I have read How to Take HDR Photos of Moving Objects? and I am asking here because I am not satisfied with what I have read to date.

When I bracket my images, they tend to come out fine but only if there is little movement in the scene. HDR allows you to capture a lot of dynamic range but if certain things move then there are ghosts. I find that deghosting tools can help if only small parts of the image move but if you have a passing car or leaves blowing in the wind the image tends to be pretty ugly.

People have suggested that I use a single RAW image but it simply does not capture enough dynamic range, and I can I capture all the dynamic range in one single exposure then the scene does not contain enough contrast in the first place so HDR is of no use.

Are there better ways to make an HDR image of a scene that includes moving objects?

\$\endgroup\$
5
  • \$\begingroup\$ Matt - thanks for the comment :) I have already read this though. There is a "small subjects with movement part" in which it recommends I use masks (I find this a little time consuming and the masked parts of the image would not contain all of the dynamic range - call me a pendant if you must). In the section "Majority of Subject Has Movement" it suggests using a single image (I mentionned in my post that a single RAW would not contain enough dynamic range). Also, someone suggested exposure blending - not bad but not perfect due to the increase in noise in the shadows... :) \$\endgroup\$
    – Ian Terne
    Mar 10, 2014 at 11:20
  • \$\begingroup\$ Did you read my answer to the linked question? Basically, the best option is going to be to manually blend the exposures. This will be quite effective if the areas containing movement fit within the DR of one exposure. See this answer for more detail about manually blending exposures to extend DR: photo.stackexchange.com/questions/3/… \$\endgroup\$
    – Matt Grum
    Mar 10, 2014 at 13:08
  • \$\begingroup\$ @IanTerne - if you don't feel that a question has enough answers, ask comments on the question or ask a more specific question asking for clarification. You can also offer a bounty on a question if you think it needs more attention (once you have enough reputation to do so.) As it is currently stated, this is a duplicate of the previous question, just asking for more answers (which incidentally, do not exist in this case, either an image is too broad for the DR of the sensor and multiple exposures are needed or the scene isn't too broad and can be captured by a single tone-mapped exposure). \$\endgroup\$
    – AJ Henderson
    Mar 10, 2014 at 13:22
  • \$\begingroup\$ I see, sorry - I am new to this exchange thing. I will have a look for what a bounty is (you're not talking about the chocolate bar are you :) ??) Someone has just given me a pretty good answer so I will post there instead. Thanks for your help:) \$\endgroup\$
    – Ian Terne
    Mar 10, 2014 at 15:30
  • \$\begingroup\$ see also this question/answer: photo.stackexchange.com/questions/3394/… \$\endgroup\$
    – Matt Grum
    Mar 10, 2014 at 17:16

1 Answer 1

2
\$\begingroup\$

There are only two ways to achieve the HDR / tone mapping with the equipment you have and you've explored both without finding a good result. The only way to achieve the better quality that you want is to capture more dynamic range in the first place.

One option is to obtain hardware that captures more dynamic range in a single frame, letting you bypass the HDR compositing step and skip straight to the tone mapping that is synonymous with HDR, that could be something as simple as investing in a new body (maybe medium format.) A pro body with a fast shooting rate may be enough to bring the ghosting to a tolerable level.

An alternative may be to look at a way capturing multiple frames simultaneously using multiple bodies. The control systems exist for stereoscopic/3d imaging. There would still be some ghosting but depending on distance to subject you might have more luck there.

Failing that you may be able to introduce an optical beam splitter into a multiple camera setup (they're used in 3CCD video cameras) that may introduce other noise into the capture but you'd need to do the experimenting there.

\$\endgroup\$
1
  • \$\begingroup\$ Hi james, I don't really have the funds for a new camera body :) so in terms of hardware I may have to look at other options. I appreciate your response \$\endgroup\$
    – Ian Terne
    Mar 10, 2014 at 15:36

Not the answer you're looking for? Browse other questions tagged or ask your own question.