How to postprocess sports?

luisferfrancoluisferfranco Registered Users Posts: 47 Big grins
edited September 16, 2006 in Sports
I'm shooting football (american football as we call it in Mexico), I'm using an alpha 100 w/300mm lens (Quantaray 70/300mm F4/5.6), I was reading a few days ago one very good post about what to watch for in a football pic, so I think this pic could be good, since it's a DB, you can see his facial expression, and how he's looking for the receiver and/or ball. What I'm not sure, is how should I process it?

Prior to the alpha, I had a Sony F828, a great camera that produced deep saturated colors, so my first try was to emule something like that, but I'm not sure if that are the adjustments that work.

Here is the first pic, it has little (very little) cropping, and rescaling for forums:
db2km2.jpg

Here is my take as producing a velvia-like photo, I'm not sure if this is correct for processing a football pic. Any sugestions will be welcomed.

db1ys3.jpg

Comments

  • stephiewilliamsstephiewilliams Registered Users Posts: 168 Major grins
    edited September 13, 2006
    Hey nice pic, very portrait like. I think the colors are too saturated personally but then again I am not much into that extra saturated look. Maybe just a subtle bump on the saturation...+2 or +3? Usually with my sports photos I apply USM, bump the sat a smudge and adjust curves and or levels if I need to. Sometimes I will blur out the background a bit more if there is something very distracting in it. Thats about it for what I do.
    Stephie
    "AMATEURS try till they get it right, PROS try till they cannot possibly get it wrong."

    Gallery - http://stephaniewilliams.smugmug.com
  • TomaSTomaS Registered Users Posts: 314 Major grins
    edited September 13, 2006
    Agree with Stephie - the second image is way overprocessed. Just look at the noise in his right forearm. The first looks fine on my monitor. Just needs some sharpeneing - the face is a bit soft.
  • luisferfrancoluisferfranco Registered Users Posts: 47 Big grins
    edited September 13, 2006
    TomaS wrote:
    Agree with Stephie - the second image is way overprocessed. Just look at the noise in his right forearm. The first looks fine on my monitor. Just needs some sharpeneing - the face is a bit soft.
    So the advice would be to take the pic as it comes out of the camera, and apply only a bit sharpening?

    What about things like contrast, or something else?

    What about the cropping? should it be tighter?

    Thanks!
  • johngjohng Registered Users Posts: 1,658 Major grins
    edited September 13, 2006
    So the advice would be to take the pic as it comes out of the camera, and apply only a bit sharpening?

    What about things like contrast, or something else?

    What about the cropping? should it be tighter?

    Thanks!

    Personally I think color/contrast decisions are best made on a case by case basis. You set a default value in-camera that suits your general tastes. But because lighting on every photo will be different and you may have to adjust levels on a given photo the saturation and contrast may change. So you may have to bump up contrast or saturation on that photo.

    As for cropping - a lot is personal preference. In my mind, for a photo where the face is the key subject I prefer the larger portion of the frame to be where the person is LOOKING. So, since the DB is looking backward my preference would be to have his eyes to the left of the frame with roughly 2/3 of the frame ahead of HIS line of site. There are always exceptions, but visually a viewers eyes tend to get drawn to the subject's eyes and then follow the subject's gaze path. So, using the rule of thirds concept, the subject's eye's should be around one of the intersection points shifted towards one end of the frame or the other with the gaze path traversing the rest of the image. So, in your image, the eyes should generally be around the left upper or middle intersection of a rule-of-thirds gridline. Generally speaking - you don't need to map it precisely. And, again, this is just a general guideline applied here because there is no other subject in the photo - things change when the photo has 2 subjects - like the eyes and a ball for instance.
  • scottkwestscottkwest Registered Users Posts: 71 Big grins
    edited September 13, 2006
    red car
    I agree with the posters above, and I would also consider cloning out the red car in the background.
  • mercphotomercphoto Registered Users Posts: 4,550 Major grins
    edited September 13, 2006
    The answer to this question will depend a lot on what exactly you are doing with these football photos. Are you shooting speculatively and hoping for orders? If so you will shoot a lot of pictures, and this will influence how much post-processing you can do. Are you shooting on assignment (i.e. the parents pre-order pictures), in which case you can shoot far fewer photos and process more later. Are you shooting for a newspaper? If so then you shoot a lot, find the few you will publish, and then do very little anyway because photojournalism ethically requires very little post-processing to photos.

    With regards to your two posted images, the second is just over-done. The first is rather nice. The crop looks good, the focus is fine. Could use some sharpening. The car in the background is fine.

    Now, assuming you are shooting a large number of photos and are shooting a league on a speculative basis, most people will shoot something like this in large-fine JPG, with contrast, saturation and sharpening boosted up a notch. I know several who dial in +1/3 a stop of exposure compensation because it helps with the inevitable shadows you get from the helmets on the faces. They do not worry about blown highlights. Another possibility is a touch of shadow enhancement in Photoshop, maybe 7-10% across the board, but you need to be careful with too much.

    While there is certainly a lot you can do on a case by case basis with these photos, doing so with 200-300 photos per game, 3 to 4 games a day, is simply not in the realm of reality. What you need to find is an in-camera setting that gets you very close most the time, and some minor post-processing that you can apply in batch that helps all photos some and hurts nearly none of them.

    One other alternative. I don't do football but I do motocross. This weekend I'm expecting to take over 1,500 photos. I will run them through an action that does basic simple adjustments and publish those as "previews". As sales come in I go back to the original and then hand-edit those. This is feasible because you can spend time hand-editing the ordered shots, but doing so to every single photo is not possible. The down-side to this is the customer does not see the true finished product online when he orders. I'm not sure yet how much a negative this is. Its not a big deal for daytime (when the results are pretty accurate in-camera) but for night racing the results from a careful hand-edit can be huge.
    Bill Jurasz - Mercury Photography - Cedar Park, TX
    A former sports shooter
    Follow me at: https://www.flickr.com/photos/bjurasz/
    My Etsy store: https://www.etsy.com/shop/mercphoto?ref=hdr_shop_menu
  • dbaker1221dbaker1221 Registered Users Posts: 4,482 Major grins
    edited September 13, 2006
    I agree with the others. The second is too saturated. merc & john make some excellent points. I think it is a very good photo though.thumb.gif
    **If I keep shooting, I'm bound to hit something**
    Dave
  • wxwaxwxwax Registered Users Posts: 15,471 Major grins
    edited September 13, 2006
    One thing I haven't seen mentioned is that face and, in particular, the eyes need to be seen.

    Helmets and caps shadow them. But by using curves or your favorite dodge method, you can lighten them up again.

    Letting us see a player's eyes will make a much, much stronger photo.
    Sid.
    Catapultam habeo. Nisi pecuniam omnem mihi dabis, ad caput tuum saxum immane mittam
    http://www.mcneel.com/users/jb/foghorn/ill_shut_up.au
  • Steve CaviglianoSteve Cavigliano Super Moderators Posts: 3,599 moderator
    edited September 13, 2006
    wxwax wrote:
    One thing I haven't seen mentioned is that face and, in particular, the eyes need to be seen.

    Helmets and caps shadow them. But by using curves or your favorite dodge method, you can lighten them up again.

    Letting us see a player's eyes will make a much, much stronger photo.

    15524779-Ti.gif with Sid. You can tweak this pic all you want. But, IMO, if you don't lighten the face, especially the eyes, it's all for naught ne_nau.gif


    I was going to wait until I got home (to my good monitor) before commenting. But, the point Sid brought up was the first thing that crossed my itty-bitty mind, when I first viewed your shot.....Laughing.gif So I just had to jump in rolleyes1.gif

    Steve
    SmugMug Support Hero
  • luisferfrancoluisferfranco Registered Users Posts: 47 Big grins
    edited September 13, 2006
    mercphoto wrote:
    Are you shooting speculatively and hoping for orders? If so you will shoot a lot of pictures, and this will influence how much post-processing you can do. Are you shooting on assignment (i.e. the parents pre-order pictures), in which case you can shoot far fewer photos and process more later.

    I'm doing a mix of both: a friend of mine is the executive secretary of the ONEFA, which is the organization for the college football league, he gave me the press access to the fields, in exchange that I provide him 24 "good" pics a week (nothing specific). So he puts this pics on their webserver, in a gallery, it's not the prettiest, but you can see it here:

    http://www.onefa.org/mayor/mayor.htm

    In the left there is something that says "Galería de Imágenes", you click here and you'll get the pics, most of them are bona fide fans which gave the pics for free. Mine are among them, but they come with my website (which is pretty empty right now) http://www.fotorecuerdo.com.

    After two or three games, the players (and girlfriends, parents and the whole family) check the pics, and they contact me to get some copies, after that I tell them that I can get pics on "assignment" (which is of course much much easier to do, and more profitable)
    With regards to your two posted images, the second is just over-done. The first is rather nice. The crop looks good, the focus is fine. Could use some sharpening. The car in the background is fine.

    So the first one is OK as it is??? really??? Maybe is just me, that is very used to see the pics coming out of a P&S (more saturated for example). I'll apply more sharpening.
    I know several who dial in +1/3 a stop of exposure compensation because it helps with the inevitable shadows you get from the helmets on the faces.

    Good idea, thanks!
    They do not worry about blown highlights.

    That's true also. Sometimes as photographers we're seeking for the ultimate and pure perfectness while our customer don't notice at all, all the effort put in the processing.
    I will run them through an action that does basic simple adjustments and publish those as "previews".

    I must learn more about Photoshop, I'm good with the tools, but I have no idea on how to program/do batch action things.

    Thank you very much for all your comments.
  • luisferfrancoluisferfranco Registered Users Posts: 47 Big grins
    edited September 13, 2006
    Thank you wax and Steve, I'll apply that, and see what results I get.
  • Steve CaviglianoSteve Cavigliano Super Moderators Posts: 3,599 moderator
    edited September 13, 2006
    Luis,
    His face wasn't as dark on my home monitor as on my work monitor. Although it is somewhat under exposed. So if you lighten too much you will bring out some serious noise ne_nau.gif

    db2km2.jpg

    I lightened the face, upped the contrast (using curves), applied some USM (.2, 500%), selectively blurred the upper background and used Alt-Ctl-~ method of darkening those overly bright areas.

    66823969.qhr8o00z.db2km2a.jpg

    Nice shot :D and if my post processing your image bothers you, just let me know and I will try to restrain myself next time :-)

    Steve
    SmugMug Support Hero
  • mercphotomercphoto Registered Users Posts: 4,550 Major grins
    edited September 13, 2006
    So the first one is OK as it is??? really??? Maybe is just me, that is very used to see the pics coming out of a P&S (more saturated for example). I'll apply more sharpening.
    Its better than the second. Is it good enough as-is? Depends entirely upon the intended buyer, what price they are willing to pay, and how picky they are. Its certainly much above amateur level as-is.
    Bill Jurasz - Mercury Photography - Cedar Park, TX
    A former sports shooter
    Follow me at: https://www.flickr.com/photos/bjurasz/
    My Etsy store: https://www.etsy.com/shop/mercphoto?ref=hdr_shop_menu
  • troutstreamingtroutstreaming Registered Users Posts: 116 Major grins
    edited September 13, 2006
    ...

    So the first one is OK as it is??? really??? Maybe is just me, that is very used to see the pics coming out of a P&S (more saturated for example). I'll apply more sharpening.

    Good idea, thanks!

    ...

    I must learn more about Photoshop, I'm good with the tools, but I have no idea on how to program/do batch action things.

    Thank you very much for all your comments.
    My two cents on using photoshop with batch processing - if you have the disk space make each of your changes on a seperate layer - and target them to push the edges, be it saturation, sharpness, curves etc. and DO NOT FLATTEN the image upon saving. Run your batch action and then go back and take a quick look at your output - if everything looks fine great, if a couple of pictures end up overdone then just open up the individual file and start backing off on the layer blending until it is more agreeable. If they are all overdone, then duplicate your action and edit the offending move and rerun your batch. Having the computer do the time consuming grunt work while you sleep, eat or shoot more is a good thing.

    For what it is worth, my batch prep is usually -run noise reduction; punch the colors with curves in LAB (then return to an RGB space); add a levels layer for setting white and black points; add a curves layer for adjusting mid-tones with curves/color balance; and add a generic sharpness layer. I will also have run, but not 'included' in the output additional sharpening layers calculated with different methods and a layer with the brightness lowered for quick and dirty burning. Then I have an action which prepares these files as jpg's that can be uploaded (resize, add watermark, convert to sRBG.) For images that I know people want or that I like and want to show off, I then go back to the batchd PSD files and hand tweak the settings and add masks to help isolate the subject etc. - but because my original run created most of the layers needed to adjust all of this, I spend less time waiting on PS to calculate what I need and can move through images as fast as my masking skills allow. The downside is that this can eat up a lot of disk space in a hurry, so I also have an action that does nothing but open, flatten, and resaves psd files, so that when I am reading to archive and move the folders off my active hard drive, I can flatten the files en masse.

    Good luck,

    Andy

    PS I hear you on the saturation and struggle with finding the line between colors that pop and those that bleed... For your image, you can always push the saturation a little (somewhere between the two examples) and then mask out everything but the uniform so that the skin tones are not blotchy and you are not calling any additional attention to the background.
    www.troutstreaming.com
    Outdoor and Sports Media
  • luisferfrancoluisferfranco Registered Users Posts: 47 Big grins
    edited September 14, 2006
    Luis,
    His face wasn't as dark on my home monitor as on my work monitor. Although it is somewhat under exposed. So if you lighten too much you will bring out some serious noise ne_nau.gif

    This thing about calibration is a real art. I use LCDs, @home, were done the postprocessing I use a so-so laptop monitor, and they looked too dark, @office where I got a much better monitor, it doesn't look that dark, in fact, I kinda like my original pic.

    About your postprocessing: man, that was GREAT, how do you do that? do you apply lightening in a separate layer to the whole image, and then mask??? I really LOVED what you did here.

    Thanks a lot!
  • Steve CaviglianoSteve Cavigliano Super Moderators Posts: 3,599 moderator
    edited September 15, 2006
    This thing about calibration is a real art. I use LCDs, @home, were done the postprocessing I use a so-so laptop monitor, and they looked too dark, @office where I got a much better monitor, it doesn't look that dark, in fact, I kinda like my original pic.

    About your postprocessing: man, that was GREAT, how do you do that? do you apply lightening in a separate layer to the whole image, and then mask??? I really LOVED what you did here.

    Thanks a lot!

    Luis,
    To lighten the face I used an action found here It's called Katrin Eismann's Fill Flash Action. I run the action, set the desired opacity (in your case I left it at 100%), flatten, select the entire image, copy, discard the action image and paste. This gives a new layer with the fill flash action. I then enabled a layer mask for this layer and erased all but the player's face. I then set the opacity and fill to about 50%. Then I flattened the layer.

    This lightened his face and eyes enough, without looking too phony (or out of balance for the lighting). I also sharpened using a dupicate layer so I could layer mask and erase the sharpening from the foreground and background.

    Layers is where it's at for me....lol I never dodge or burn (I don't even know how to rolleyes1.gif ). I just apply the appropriate type layer and erase or paint clap.gif Yes, I am old and lazy. Not to mention 1/2 blind....lol

    The Alt-Cntl-~ method works like this. Once you hit all three keys together, the highlights of your shot will be highlighted by those marching ants (those little moving dots). Select Edit in Quick Mask, Image>Image Adj>Levels and move the left slider at least 1/2 way towards the right. You will learn to judge by what the histogram is showing. Then select Edit in Std Mode from the toolbar, Ctl-J which will bring up a new layer and select Multiply as your blending choice. Toggle the new layer on and off to see the effect. You can lower opacity if it is too strong and/or you can use a layer mask to erase the areas that might be too strong while leaving other areas un-touched. This method can be used multiple times if you really have some major glare to subdue.

    Hard to do a tutorial on the fly, so if this doesn't make sense to you. Email or PM me and I will try to send you a better explanation of the workflow thumb.gif

    Stvee
    SmugMug Support Hero
  • luisferfrancoluisferfranco Registered Users Posts: 47 Big grins
    edited September 15, 2006
    Thanks man, I'll see what can I do this weekend.

    Yesterday was my first approach with dodge/burn, looks like a promising tool for me, but my results are too fake yet. I have to learn how to control this. I think there are several ways to kill the same animal. I'll try your method, and let you know if all the mask stuff is easier than the dodge/burn or otherwise.

    Thanks a lot!
  • BBonesBBones Registered Users Posts: 580 Major grins
    edited September 16, 2006
    fahk Steve, I need to learn some of those Mad l33t skills. Nice touch up job!!!
Sign In or Register to comment.