How do you select the 10 best images from a set of 300?

Question

I have read that photographers are rarely the best editors of their own images. As digital cameras make it so easy to take hundreds of shots of a given subject, the challenge of sorting and ranking them to come up with the handful of great images to share/print/publish has become so monumental that it is almost hopeless. What tools and techniques do you use to perform this task of curating images?

Answer 1

Here's some of what I do:

• Throw out the worst first. Blurry, blown out, excessively dark/noisy. Back-of-somene's-head is usually included here too.
• Repeat the above rule a few times, raising the bar for "worst" so that it's relative to the new set.
• Try to eliminate duplicates. This is an especially big deal when shooting in burst mode. Take X pictures that are very similar, keep exactly one. Picking the best 1 out of 5 is easier than picking the best 10 out of 50.
• Do a very brief pass over the remaining pictures, and assign some sort of rating (usually 3-5 stars by this point) based on a very brief impression of the picture. You're looking for overall aesthetic impression here, and I find that it's best judged at "first impressions". Don't think about this too much.
• Depending on how you rate your photos and how many you have, you'll probably only want to share/publish a fraction of your highest-rated photos. Less is more; even a dozen really good photos from the same scene can get boring. Throw out anything that's at all uninspiring.
• Don't get bogged down by technical qualities: framing, exposure, contrast, colour, blemishes, depth of field. All of these can be fixed in post-production if it's worth spending time on. Do pay attention to anything that can't be fixed in post: focus errors, posing.

Basically, the idea is to get really aggressive with throwing out pictures, and then spend time editing the keepers to make them awesome.

Adobe Lightroom makes this very easy; you can "reject" pictures (flag them as "bad") with one key. If your filter is set right they'll disappear immediately from your working set. You can also change your filter to include them; nothing is ever lost/deleted.

Getting good at culling the photographic herd means that you can let yourself take more shots on-site, which in turn increases your chances of getting that one amazing photo.

Answer 2

Not sure that a picture with incorrect histogram, excessively dark/noisy or blurred should be removed immediately. Sometimes I see that even defective picture looks good after time. For example (as for me):

So I found the best method to select the best pictures: I just show the pictures to my wife.

Answer 3

Delete the bad ones vs. Keep the good ones.

Until some months ago I have always taken the usual approaches: Mark the ones which are not really good, delete them, and repeat this step multiple times. I found this was very time-consuming and at the end I still had a lot more pictures than I wanted to have.

My new way is the opposite: Mark the images you want to keep. After quickly going through all images to get an overview, I only tag the images where I really think: Wow, this looks cool, I have to keep this one.

This has two advantages: The first being that you can simply ignore the worse images. You spend extra effort on the good images instead of on the bad ones (tagging). The second advantage is that with the usual approach I tend to have lots of images like «yes, it is sharp … and the special thing about it is?». I.e. it is not a clear candidate for deleting, but not interesting either. These are the images I won't take a look at later anyway, and which don't pass the second approach (keep the good ones).

Also it's faster and requires less iterations :)

Answer 4

Since some people in the question comments repeatedly told me it was rather trivial, I actually did it: An application that sorts the images in the current directory by leaving the actual head-to-head comparison to the user.¹

Written in C# for .NET 2. Works on Mono (tested on Linux so far), too. Requires dcraw on the PATH (compiled executable for Windows or OS X can be downloaded here).

Currently the user interaction is very rudimentary. This is probably subject to change. Also the code is an evil mess as usual with such a hack-job.

The application has to be started in a directory that contains the images to sort. It then proceeds to load all images it can find and handle (JPEG, GIF, PNG, BMP are supported natively, all other formats are either supported by dcraw or skipped if not). Keep the number of images reasonable, as every image is preloaded in memory to speed up display – I tried starting it on a folder with around 600 images and terminated it at around 2 GiB memory usage.

After that you get a two-pane view with an image on the left and on the right. Click on the one that you consider the best of the two. You will then get two new images. Continue until done. You can close the program if you want, it will resume where you left off.

After all necessary comparisons have been done² the result can be seen:

It has the sorted list of images on the left with the highest-ranked at the top and the lowest-ranked at the bottom.

Todo list:

• Allow selecting the images to sort.
• Solve the issue that portrait pictures are always displayed in landscape orientation (at least for raw images. dcraw allows rotation but cannot do so automatically and I don't see an easy way of finding that out externally).
• Reduce memory usage for large numbers of images.
• Shuffle the images beforehand so bursts of nearly identical images are less likely to be compared directly against each other.
• Change synchronization between the sorting thread and the UI to no longer rely on Thread.Sleep and polling but to use proper synchronization methods.
• Add a 1:1 preview (or at least a larger one). Currently this cannot be used to judge things on the pixel level.

It's 5:26 here, so I stop hacking on that now.

Source code can be found in my SVN repository and is released under the MIT license. I welcome patches ;-)

Images in above screenshots are my own.

---

¹ Of course it wasn't as trivial as others would want to make me believe. After a long struggle with Libraw, I simply went the dcraw route. Not pretty, but works with minimal amount of code.

² This is in the order of n log₂ n where n is the number of pictures compared – so, for 20 pictures you can expect something around 20 × 4.3 ≈ 85 comparisons – I know, it's not a small number. For the 300 images you mentioned you'd get around 2400. The actual number that has to be performed manually is (a) different (since complexities omit the linear factor) and (b) as far as I observed so far, smaller. To avoid inconsistencies the user will never be prompted twice on the same two images (either order) and never prompted with the same image on both sides.

Answer 5

This article by Chase Jarvis explains his awesome workflow. He shows how he selects few pics to put on his portfolio and show to his clients from a set of about 15K clicks..

This is the summary: There are 5 stars available in any workflow management s/w like bridge or aperture. He uses the stars in following manner to sort the pictures out.

1* – Pace: Full Speed. Main Criteria: Is it garbage?

2* – Pace: Full Speed. Main Criteria: Does it look OK?

3* – Pace: Cruising. Main Criteria: Is it pretty nice?

4* – Pace: Calculated. Main Criteria: Is it nice when you look closely?

5* – Pace: Slow and thoughtful. Main Criteria: Is it outstanding? These are the winners.

You should really read the article.. it has really helped me alot!!

Answer 6

I have struggled with the same question on occassion. One of the first things to look at is how you are reviewing the images. There are various tools out there that can help with the task. For example there is Lightroom and PhotoMechanic to name two of the popular ones. These have rating tools, or flags for setting Pick or Reject. However this task can also be done with things such as Windows Picture Viewer and Macintosh Preview.

The key I have found is to do a complete backup of my images before I start culling the images. The reason I find that to be helpful is that after that I no longer stress about, what happens if I reject a keeper. If I do, I go to my archive of the image and bring it back.

The other thing I do is not to look at all the images at once when starting. I look at one image at a time, so that I do not know if there is a similar image next in the collection I am reviewing. I also don't get overwhelmed. I quickly discard images, just simply out of focus, glare, not framed well, someone blinked, plane in the sky... etc. I go back and do it a second time as I have a little more perspective.

On the third time through I go back through and look at the whole gallery to start seeing if there are images I can group together to compare to one another. I do not try to compare three images to each other. For example if I have three images, A, B, C I do not look at all three. I pick the better image between A & B. I then pick the better image between the AB Winner versus C. I find it easier to just compare two items to each other than to try to look at three or more as the variables start to get larger.

I see where my count is after that first "grouping" through and repeat as needed.

I also take breaks between the review sessions as after a while objectivity gets lost. Everything starts to blur together and fatigue sets in for me.

Answer 7

I have consistently found that an iterative approach leads to the best results. How you do the iterations depends on many factors, including what you mean by best, the % of good and bad photos you take, how many of your photos are technically ok (in focus) etc.

Either way, the goal is to eliminate more photos quickly in the early passes and pay closer attention to each photo in the later passes. In the first pass I eliminate maybe 75-90% of my photos (the obviously unusable ones, or the ones that are too similar to other photos). In the second, I start cropping some photos to get an idea if the composition works well. In this and later iterations I start rating my photos to get a general ranking. In the second or third I will start doing basic retouching. On the last couple of iterations the idea is to spend a lot of time on a few photos.

Answer 8

The best for what, is always the question.

If something is going to be printed large or shown where you want people to admire both technique and artistry, then you start by removing technically flawed images.

If the image is going to someone who is not going to be primarily judging on technical merit, you probably want to look for the best composition or expression and lean on tooling to help recover an image that is technically flawed.

Often you have to step outside yourself, and think about why someone besides yourself may like an image for reasons you would not consider personally important.

Answer 9

One of the toughest, but most effective, things to do is discard (I don't necessarily mean delete forever, just don't include it in that final 10) any image that isn't technically perfect. So if you didn't nail the focus, or there's something distracting in the picture (whether it's litter or you didn't frame the picture right so there's a tree growing out of the person's head), or the picture wasn't level, or any of a dozen other things that can go wrong, then don't share/print/publish the picture!

It's tough because it means discarding a lot of pictures that are otherwise great: you perfectly captured the light, or the look in the person's eyes, or whatever else. So share the picture with your friend who was there with you, or with the subject or their family, and they'll enjoy it, but don't share the picture with a wider audience because anyone who wasn't there or doesn't know that person will just be distracted by what's wrong with the picture and won't enjoy it as much.

Answer 10

There are some great answers so far, so I won't reiterate what others have stated. On a different level, I think it should be stated that there is not necessarily one correct, "algorithmic" way to do this. This process is one where the human mind truly excels, and where rigid algorithms and "computers" break down.

Choosing the top 10 "best" photos from a set of 300 is a very subjective, even emotional process, and one best suited to the individual. What may seem like a superb photo to one individual may seem like crap to another, and the factors that go into making such a judgment for each person are going to differ, possibly even be diametrically opposed.

I think the best way you can help yourself figure out what images to keep, and what images to throw away, is to develop your sense of vision and style. Vision is your ability to "see" into a scene or photo shoot, and visualize what you want to get out of it. Style is formulating your own personal sense of aesthetic appeal and a pattern of consistency (or, even, lack thereof.) If you know what you want when you sit down to shoot, and understand what is aesthetically appealing to you, filtering through 300 photos to pick the top few that are keepers becomes a much simpler job.

It'll be an entirely personal process, one built on your own outlook on life, and possibly incompatible with many other people...but it will help you in more ways than simply picking the photographs you want to keep from a shoot. It'll ultimately help define who you are as a photographer in the end.

Answer 11

Your first hurdle here is to define what you mean by "best" in the context of that particular set of images.

What's the purpose that collection of 10 is going to be put to? Is it for submission to a photography course where they'll be judged on technical merits (and which?)? Is it a wildlife shoot for National Geographic? Is it for a wedding album of your niece? Or maybe a wedding of a paying customer? Is it simply vacation snapshots of you and your family? Is it a submission to a themed competition? Are they to be added to your personal portfolio to be used as a marketing tool for potential customers in any of the above or a dozen other categories?

Each of those (and think up your own) scenarios will require a different (but maybe partially overlapping) definition of "best". Some of those criteria might be very easy to judge, some almost impossible without serious thinking about each image.

Personally, I have typically no problem throwing out 50-75% of a shoot in a first sorting. After that it gets harder as images now conform to the purpose of the shoot, and have to be compared to select which is of higher technical quality and/or better subjectively (and worse, has the best overall combination of the two, a technically inferior image might be so much more suitable for inclusion than a technically superior one on emotional or otherwise subjective criteria it does win out in the end).