PDA

View Full Version : What's your digital workflow?


ZiggyB
08-23-2005, 09:07 PM
So I'm wondering what other people do when they get their photos off of the camera and then to the final product, being a web page, print, etc?

This is what I usually do after coming back from a convention and put photos online.

o Copy all the photos from digital media to my laptop.

o Make a backup immediately to DVD-R (Using multisession so that I can add my fixed images later)

o Use ACDSee to view all of the pictures at a glace and delete the ones I don't want.

o When I pick out the photos I do want to show, I have it loaded up into Photoshop CS2.

o Tweek the levels and sometime bump up the contrast +10. I've noticed that it tends to make "punchier" images. But this was before going to DSLR. I've noticed that I've had to do less contrast adjustments from photos taken with a DSLR.

o Save a copy of the "adjusted" image, still keeping the original around.

o Resize the image for the web, 800x600 for my web site, 640x480 for Cosplay.com

o See if it needs a little bit of sharpening. Since reduced images tend to lose a bit of shapeness. I'm still learning how to use Unsharp Mask, so a lot of times I get lazy and just use the regular "Sharpen" filter.

o Add a watermark image (that points to my web site) to the image.

o Use the "save for web" function to save the web ready images.

o Upload the images to my web site and images.cosplay.com.

After I've made adjustments to all of the images, I then burn all of the changed images and resized images to another session of the DVD-R that I started with.

Be careful though, DVD-R multisession, apparently does not work with Windows 2000 or below. I'm not sure how Macs handle multisession DVD-Rs.

What do other people do?

Godly
08-23-2005, 10:05 PM
Hooray for imaging thread. Exchange of tips and tricks I see.

1) ACDSee photos I want/don't want
2) Open up Photoshop (6.0)
3) Rotate/Crop
4) Lighting/Contrast editing (I use Curves)
5) Color Correction
*6) Reduce Image size to 640pixels *Will start on next opportunity
*7) Unsharp Mask Filter *Will start on next opportunity. Just heard about it from tipsfromthetopfloor from other Ziggy thread. Did trial run on photo, loved results.
8) Add watermark to top corner
9) Flatten Image
*10) Save jpg. Trial and error quality to meet file size restriction *Will start 'Save for Web' on next opportunity after reading Ziggy's imaging process.
11) I keep copies of the raw files on my HD.

Ani-Magic should be fun for me. Learn something new every con.

EDIT: By raw file I mean original jpg. I don't have the option to not save in jpg.

quadrain61
08-23-2005, 10:58 PM
1) Copy photos to two locations (regular photos and a random 'New Folder' on the desktop for editing etc.)
2) Create folders named '1', '2', and '3'
3) Open 'Microsoft Office Picture Manager' to see all my pitures and see which ones i want and don't want in the 'New Folder'
(all in Picture Manager from 4-6)
4) Resize ALL the pictures to either 640 pixels
5) Rotate which photos needing to be rotated (since my camera doesn't do that automatically like the new newer cameras out there (I.E. canon digital elph)
6) Go through EACH photo and adjust lighting or anything basic
7) Go to the 'New Folder' and put the 640x480 in folder '1' and 480x640 in folder '2' for preparation of the power of photoshop
8) Open photoshop and run a batch for either folder '1' or '2' depending.

I recorded actions (batch) in photohop to automatically
-Open each image in the folder
-Place a watermark on the bottom right corner (this is the part where I had to seperate the two folders because to place the watermark I had to specifically tell it where pixel wise)
-'Save for web' (lowers the quality more to save space) and save the finished file into foler '3'
-Close
-repeat (until it does it for every file in the folder)

9) Review folder '3' to see if something's wrong
10) Upload the photos using the 'Gallery' (http://gallery.menalto.com/) application on my site so it can automatically thumbnail and such.

11) Back up the raw (the same definition Godly gave) files to CD(s) or DVD

It's not that I'm lazy to edit photos in photoshop, but it's the time. During the non-con days I'm constantly at school and work.

From reading Ziggy's and Godly's posts, I guess I need to check out the ACDSee program.

jtnishi
08-24-2005, 01:12 AM
My workflow goes something like:

1) Copy from main location to subfolder for working files (I handle archiving in a separated workflow - I burn DVDs off by convention).
2) Use ACDSee to select & rotate images.

Within Photoshop:

3) Select (fixed ratio 4:3 or 3:2, depending on original) & Crop (recompose)
4) Touchup work - removal of skin blemishes, lightening of dark circles under eyes, etc.
5) Resize to either 900x600 or 800x600 (I do the resize step early because I normally do web copies & print copies in separate runs, and I need speed on my side)
6) Auto Contrast
7) Brightness correction & color correction in curves (gray dropper trick + midpoint curve manipulation for midtones)
8) Density correction using either curves or the new shadow/highlight tool.
9) (secret weapon) OptikVerveLabs virtualPhotographer Filter, film#2 or slide#1 mode, warm film emulsion (between -12 and -5 typically) - increases contrast and adds that perfect tonality for those outdoors shots. I started using this since Ani-Magic last year, and it added that last nice touch to the shots that gave it that film look I've always wanted.
10) Unsharp mask - 85-115%@0.5-0.7px on average, stronger as needed, faded to luminosity.
11) Save for Web -> the larger of 50 quality or 120kb.

12) Generate thumbnail

The rest is all administrative. Now if we could only figure out what Maboroshi's automatic workflow must be like... ^_^

ZiggyB
08-24-2005, 01:33 AM
Godly: Yah, I've always avoided sharpening because I've noticed how bad it can look when people over do it, but when done right, the change is quite noticable.

The nice thing about "Save for Web" is that it takes out a lot of the trial and error. It will show you what the compression quality will do to the image. Just slide it from like 10% to 90% and you can see what I mean. It also reports the size of the final image. One interesting bug that "Safe for Web" fixes for some versions of IE. It seems IE choaks on EXIF data in images sometimes. Safe for Web removes that data.

I have RAW on my Canon, but I haven't used it yet. Right now, the JPG is fine and I don't want to deal with the extra step of converting the RAW and having it use up more of my memory card. But I know plenty of people who refuse to shoot in anything but RAW. :)

quadrain61: You can save yourself a step or two with your watermarking by using Layers->Align in Photoshop. This way it will just flush your watermark image without you having to give exact pixel coordinates.

As for ACDSee (http://www.acdsystems.com/English/index.htm), it's just an image viewing program. It does come with some basic editing (like batch resize and such) and some good cataloging features. However, I like it mostly because I can browse my pictures and then when I want to edit, just click a button and it loads in Photoshop (or any graphics program actually). It is a pay package though. I've been meaning to try Picasa (http://picasa.google.com/) from google. I think it does similar things and it's free. But I've been lazy and I'm too set in my ways. ;-)

jtnishi: I'll have to try those USM settings. I'm still learning it and I'm desperately trying not to oversharpen.

Yah, I would also really like to know what Maboroshi's workflow is. This is just my personal opinion though, but his photos looks a bit too over processed for my tastes. Though that maybe just his style and you have to admit, when you see one of his photos, you just know it was taken by him. :)

jtnishi
08-24-2005, 11:05 AM
jtnishi: I'll have to try those USM settings. I'm still learning it and I'm desperately trying not to oversharpen.

Yah, I would also really like to know what Maboroshi's workflow is. This is just my personal opinion though, but his photos looks a bit too over processed for my tastes. Though that maybe just his style and you have to admit, when you see one of his photos, you just know it was taken by him. :)

Yeah, those settings seem to work for me. What you might try is that in the fade to luminosity step of the unsharp mask (after using USM, it's Edit->Fade _), you can lower the percentage of the effect of the USM.

As for Maboroshi, yeah, I've always thought his pics looked a bit overprocessed too, but I think I remember reading somewhere that his process is also completely automatic. And frankly, I've been looking to try to automate my process much more since at the moment, I average about 2-4 mins per picture to edit, which is simply turning out to be too much time.

quadrain61
08-24-2005, 11:24 AM
quadrain61: You can save yourself a step or two with your watermarking by using Layers->Align in Photoshop. This way it will just flush your watermark image without you having to give exact pixel coordinates.

As for ACDSee, it's just an image viewing program. It does come with some basic editing (like batch resize and such) and some good cataloging features. However, I like it mostly because I can browse my pictures and then when I want to edit, just click a button and it loads in Photoshop (or any graphics program actually). It is a pay package though. I've been meaning to try Picasa from google. I think it does similar things and it's free. But I've been lazy and I'm too set in my ways. ;-)

Thanks!

Ollie
08-24-2005, 12:02 PM
1) Run Automator script to change from 100_xxxx.jpg to con_yyyy-mm-dd-hh-mm-ss_ID.jpg. This allows me to take picutres on different cameras and cards without having to worry about getting the right sequence when downloading. Makes it easy to check where I was and what I was doing at the time.

2) Backup to DVD, before any kind of editing has been done.

3) Run a batch process (Graphic Converter -- does a better job at this than PS) to correct levels, apply a semi-conservative unsharp mask filter, and resize down to 2496 x 1664 (8MP images give a bit too much detail, especially with closeups).

So far, I've not even needed to look at any of the pictures yet. (Actually, step 0 is just to review them, since I'm curious.)

4) Look through the processed images. Remove any duplicates or bad shots, and make note of images which could use or deserve a manual treatment.

5) Make said adjustments to particularly good photos. Maybe 8% or so of pictures are particularly good -- meaning I got lucky with lighting and so forth. Instead of just trusting the automatic procedure, I replace these images with the originals, and rework them.

6) Rework select images. Manually adjust levels, colors, hue, saturation, and any other property which may improve the look. If needed, blend different layers together to correct separate parts of the images. Whiten eyes and/or teeth, edit out any other slight blemishes or imperfections. Apply an Unsharp Mask and crop, if necessary.

7) Scale down and save.

8) Add captions to a database, then process with JAlbum to create a webpage. Upload page to servers. And hope people like what they see.

I'm not good at judging image colors, so I tend to botch up a bit, but I usually like the results better with a bit too much color and vividness than too plain and flat.

ZiggyB
08-26-2005, 04:22 PM
Oh speaking of sharpening tricks, I read about this method for sharpening images:

http://www.tipsfromthetopfloor.com/forum/viewtopic.php?t=148

It sounds like a lot more work. :/ I'll have to try it out to see if it's worth it though.

dmk26
08-26-2005, 04:54 PM
I really should do a bunch of the steps that you guys do, except I'm too lazy =). For example, I haven't backed up my photoshoots over the last year ! I also don't do a cursory scan through of all the pictures and delete the ones I don't like...
Anyway, here's my standard workflow:

1 - Copy all files off CF card onto harddrive
(The following is done in Photoshop in groups of 3 - 4)
2 - Rotate where necessary
3 - Modify contrast or levels if necessary - For each shoot, I end up with a standard modification that I then save as an action and then use the batch workflow to go through remaining pics.
4 - Move the pictures I like into a "good" folder
5 - Go through the "good" folder for a second look through and delete remove pictures I don't want to use
6 - Resize, add border and save for web all "good" pictures.

7 - Hand over a CD with all images to whoever the shoot was with.

gmontem
08-26-2005, 05:30 PM
- Copy all the directories from my PSD to the hard drive
- Leave the room and have a :drink:
- Move the files in the HD into folders according to date
- Fire up Capture One LE and that Canon program whose name I can't remember, I think it was called EVU
- While C1 generates thumbnails and such, get another :drink: or do some :chatah:
- Swear at the computer and make death threats if the process is taking too long
- When done delete the images I don't like
- Verify camera profile is Magnes's low saturation profile
- Adjust black level and EC to my liking
- Now comes the fun part, sharpening! Oh joy! (not):
. View image from EVU with used focus point shown (C1 doesn't do this! :()
. Switch back to C1 and set zoom window to that area
. Move sharpening slider to ... sharpen! Often times I gotta move the slider back and forth like 20 times to make sure I didn't overdo it.
. Curse at myself if the image is soft from camera shake, subject movement (candids), etc. and delete it
- Set crop rectangle and mark for batch processing into a 16-bit TIFF file, camera profile embedded
- Repeat for next image
- Once all or a certain number of TIFFs have been generated, I close C1 and open Photoshop 6.0 to resize the image (the LE version of C1 won't let me do this)
- Set to Lab mode and do one final and light USM, just a teeny tiny bit
- Run the generic-looking frames action I recorded. I can't record relative coordinates with PS 6 so at one point the action pauses for me to manually move my name and the event name to the appropriate places
- Convert to sRGB colorspace and save as JPEG
- Once I have all the JPEGs created, close PS and fire up an old copy of Breeze Browser I bought back in the Canon G1 days :heart: :heart:
- C1 embeds a mangled exif in the generated TIFF file so I use BB to copy the exif from the original RAWs into the JPEGs. I like to leave the exif in the image because some people like to know what settings I used for a particular exposure.
- Curse at pbase for misinterpretting the exif data and thinking I shot in P-mode.
- If it's getting late :crest: and get some :sleep: , otherwise :waaaah: because there's still more work to do

raye
08-26-2005, 07:50 PM
- Copy RAW files to local drive.
- Convert RAW files with settings for thumbnails. This creates a small jpg of all pictures for easy review.
- Copy RAW + JPGs to backup location and to primary storage location.

- Select files for further postprocessing (selecting the files that do not obviously suck).
- Open RAW file in RAW convertor (Nikon Capture for me).
- Adjust exposure, color temp., etc. Save as TIFF for Photoshop postprocessing.
- Open file in Photoshop. Apply adjustment layer for levels, curves, channel mixer. Remove noise and sharpen if needed (Neat Image and Focus Magic are my preferred tools. Might want to try Focal Blade too (Focus Magic doesn't like 16bit images. Did not like Noise Ninja).
- Resize for print output size and flatten.
- Resize for web output and upload to webpage.
- Make prints at the local Fuji Frontier or abuse the Xerox Phaser at work.

Actions and adjustment layers make the process much faster. Doing postprocessing without automation is a bad time.

Sharpening is probably the thing I most need to get right. I was reading this tutorial (http://www.thelightsrightstudio.com/DigitalDarkroom/Tutorials/PutAFineEdgeOnYourSharpeningSkills.pdf) and it hurt my head. It is just so involved.

Celebrimbor
08-27-2005, 12:07 AM
-Import from Sony DSC to Tirion via USB
-FTP backup copy to photo archive folder on Gondolin
-Edit on Tirion with Photoshop CS
--Usually involves a quick crop and brightness adjustment
-Save for web
-Kerberos FTP to public.iastate.edu/~cverlo

tfcreate
08-27-2005, 06:24 PM
Much the same as with film.
1. Develop then make up contact sheet.
2. Archive film/data as TIFF, JPEG and EPS (to DVD/CD-ROM.)
3. Send contact sheet to subject with request for their selection.
4. Process selections (edit and clean-up.)
5. Archive finished negs, burn copies for customer.
6. ALWAYS tell customer where they can get more copies if needed.
TFC

lavender_mists
08-29-2005, 01:53 AM
Wow, I seem like quite an amateur after reading these previous posts. I have been an avid fan of digital photography and have taught students the basics (and then some) of Photoshop and Illustrator even before I got my own digital camera. But anyway, I donít think I have many steps when it comes to this whole process, even though I find that it takes an insane amount of time.

1.) Copy images from the card to the computer

2.) Use Irfanview to view and weed out the undesirable images

3.) Start with the typical cropping, colour correction, level adjustments, etc.

3a.) ZiggyB, like you, Iím starting to use the Unsharp Mask as well. I discovered this little trick to make the colours in photos a lot more vivid. I use the filter to change the levels to the following: Amount: 30%, Radius: 60.0 pixels, and Threshold: 10 levels.

3b.) I am a fan of Gaussian Blur. Not to sound mean, but I find that Gaussian Blur is the most abused filter on the net. It can definitely make or kill if not break a picture in my opinion. But all angry rantings aside, I like to use that filter to draw focus to the subject of the picture, and to make the personís skin smoother while making their other features (Lips, eyes, hair, etc.) stand out.

After a few good hours of staring at the screen and am satisfied with my editing job, hereís the part where I start getting confused and clueless.

3c.) Flatten and resize. Resizing, thatís the part that gets me the most. Is there a common ďstandardĒ that most people use? I just keep on resizing until I am just below the size limit that the servers let me upload at. I keep two sets, one that I have resized so that it can be on the net, and then the large edited ones just in case I ever feel like touching them up again.

4.) Upload and/or print away. Iím trying to look for some nice and new C1S to print some of my better pictures on. Are there any suggestions from the crowd?

And oh, for tips and tricks for photography (digital and film) go to http://www.popphoto.com. It has a lot of great information about practically ANYTHING that has to do with photography.

ZiggyB
08-29-2005, 04:03 AM
For me, resizing for the web I usually stick with 640x480 or 800x600 pictures.

However, since getting my DSLR, the aspect ratio is not exactly square, so I've started going 720x480 and 900x600.

For cosplay.com, there is a size limit of 640 on the wide side if you're uploading into the photo galleries here.

TomodachiFriend
08-30-2005, 11:54 AM
My workflow changes depending on the target audience and the quantity of pictures taken. Wedding shots get more attention while cosplay shots are mostly automated for example. Here's what I do for cosplay:

1- Transfer images to computer. I shouldn't even have to mention this.

2- Open up XnView (I used to have ACDSee too but it's too bloated yet still manages to lack some features I want) and copy the good pictures, the ones I think might be salvageable, and the ones I think are too important to be left out (a bad picture is better than none).

3- I run all of the copied pictures through a script in XnView to rotate if needed, rescale, add a watermark, and generate thumbnails.

4- Launch Photoshop to work on those pictures I thought were salvageable using the originals. If they're not good to go after a few minutes in Photoshop, I scrap them.

5- Edit a template and run an ASP script on the web server in the basement.

6- Write descriptions for every page.

7- Upload the generated pages and pictures to my web site.

Ollie
08-30-2005, 02:11 PM
I am a fan of Gaussian Blur. Not to sound mean, but I find that Gaussian Blur is the most abused filter on the net. It can definitely make or kill if not break a picture in my opinion. But all angry rantings aside, I like to use that filter to draw focus to the subject of the picture, and to make the personís skin smoother while making their other features (Lips, eyes, hair, etc.) stand out.

I use this trick, too, in a slight variation. Make a layer that's a bit over sharpened, and one that's blurred again with a gaussian. Merge the two layers in a ratio (or opacity) which looks right. It'll smooth skin without losing the difinite lines of "natural" features.

3c.) Flatten and resize. Resizing, thatís the part that gets me the most. Is there a common ďstandardĒ that most people use? I just keep on resizing until I am just below the size limit that the servers let me upload at. I keep two sets, one that I have resized so that it can be on the net, and then the large edited ones just in case I ever feel like touching them up again.

Since most people are running at at least about 1024x768, an image scaled down to 800 pixels on its longest side will fit without too much scrolling. It's still plenty big to see detail, but small enough to transfer quickly. Those on larger monitors won't need to scroll at all, but won't have a tiny image on their screen.

AgentSakur9
08-30-2005, 08:07 PM
Hmmm... I am not as organized as you guys...

1) I create a folder on one Hard Drive.
2) I upload the pictures to said folder.
3) I duplicate the folder on a second Hard Drive (where my site is Mirrored on the HD)
4) I use ACDseeIT Classic to view my pictures on that second drive
5) I delete "unpostables"
6) I then open ACDSee 4.0 and use it to resize, then rotate the pictures. (640x480)
7) I use Picture Shark to batch watermark my pictures
8) I ftp to the web server
9) Coppermine does the rest
10) Once I month I back up all the pictures onto a External HD. It then goes back into storage after back up.

I still don't trust optical media (DVD or CD) for stable backups. I have a few "backup" CD's that have degraded over the last few years... and they did nothing but sit in an air conditioned environment. So I just bought an External HD for backup purposes. Also I have the pictures stored over two Hard Drives on my computer as well. Although they need a bit of organizing.

I rarely do any photoshop work, mostly because I am lazy. If I do end up using photoshop it's mainly for prints. Removing blemishes and that sort of thing. Otherwise I try to get my desired effect on the camera itself. ^_^

ZiggyB
08-30-2005, 08:17 PM
gmontem: Hahahaha, you must have the most laid back workflow I've ever seen. ;-)

AgentSakur9: Why do you use ACDseeIT classic and then switch to ACDSee 4? Just reading it, it seems a little odd to me. O:)

As for reliable media, yah that's an interesting topic all in itself. I read (I wish I still had the link) that the most reliable, long lasting and cost effective backup method is still tape. Hahah. :) Anyone got a DAT drive to spare? :>

AgentSakur9
08-30-2005, 08:24 PM
Classic Uses less Memory in terms of RAM and loads faster (Thumbnails and for it to Initialize). It's basically a bare bones version of 4+ .... I use it for all my file managing though. I like it's structure better than Windows Explorer. Plus if you do a "right-click copy" it copies the location of the file C:/Pictures/Whatever.jpg which is handy when you're uploading to a site ^_^... no more browsing for the picture.

Yeah... like I said, my way is pretty unorganized and nonsensical.

ZiggyB
08-30-2005, 08:27 PM
Actually the latest ACDSee does that copy thing too. It's really nice, when you copy to a directory or folder it knows to copy the file. But if you copy into a text field, it knows to put the file path.

Why hasn't anyone done this before!? :) I think that's one of the biggest reasons I'll have a hard time getting away from ACDSee. :)

TomodachiFriend
08-30-2005, 08:35 PM
I don't know if I'm getting off topic here but since many have included back-ups in their workflow, I'll just add my own opinion gained from my experience. Optical media is very reliable if you buy good products. This not only includes the blanks but the burner too. I have 10+ year old CD's that still read fine even if they're kept in the hottest room in the house. On the other hand, I always had to reburn all those cheap CD's I bought. I'm sure my DVD's will hold up as well as my CD's. They'll last at least until the next generation of technology becomes affordable and I move the data to those new discs. Oh, and since DVD and CD blanks are so cheap, why not make 2 copies or more?

shiroin
08-31-2005, 04:30 AM
when i shoot i shoot RAW+Basic
so right after i shoot i can view the photos really quick and see if i should use any adjustments (yes i carry my laptop with me :p also helps when out of space, what i get for shooting RAW. I shoot RAW for the purpose of fine adjustments :))
anyways after i just copy to my HDD, and sometimes ill back it up on externals drives

since most shots are snappy, most stay in my drive for my personal amusment
others are displayed in three ways:
-snapshots for blogs, journals - snapshot grade
-photo albums - average grade
-extra processing and refining + description write up - the photography grade (here u go, why i shoot RAW)
i wont go thorugh the detials of post processing
but generally...
snapshots are resized to 750*5ish (depends on the camera)
photo albums 1000*665
photography 1060*725 (with boarders)

examples
snapshots: http://gallery.orbital7.com/Japan_Snaps/temple2
photo albums: http://r-pai.com/Japan/DSC_14461.jpg
photography: http://www.deviantart.com/view/22102447/

SpookyElectric
09-09-2005, 06:05 AM
If your scaled images don't seem as sharp as they should be, it's probably due to your software's scaling method.

As you can see here: http://www.dylanbeattie.net/magick/filters/result.html (of most relevance is the Blur = 1.0 column) there are actually a lot of ways to scale an image, some much better than others. A lot of software just does linear or bicubic scaling, which are actually not very good conpared to many other methods. Lanczos seems to work really well, and is getting more widely adoped, but far from universally.

Those tests were done with ImageMagick, which is normally used from the comand line for batch image processing. ImageMagick also lets you set a support value (blur) - if you set it to something less than 1.0 (but probably > 0.7 to avoid making things worse) you can probably get a little extra percieved sharpness.

That said, unsharp mask is a nice psychovisual effect, and it's probably still useful to add some "punch". Though I think should be able to apply it just as well earlier in the process (you'll probably need a larger radius), then you can do batch resizing + watermarking to save time.




Anyways, my process:
- Get the photos in some digital format (until last month I've been using film, so that means using someone's scanner, or getting pictures on CD)
- Crop them - I take pictures in a bit of a rush to not take too much of the cosplay's time, so they're usually badly framed. Also, I may crop out shoes that don't fit the costume, out people/objects that I think are detracting too much from the subject.
- Adjust levels (seems to be less necissary with my digital camera)
- Clean up any visible dust and other issues from scanning if they affect the subject.
- If the red-eye is bad, try to clean that up.
- scale manually: With scanned images, sometimes I make them smaller to hide some scanning deficiencies or bluring. Alternatively, sometimes I'll want a larger image for a detailed costume that I want to show off.
- save with a descriptive filename.
- upload

All processing is done in The Gimp. With my digital camera, I filter out unsable images with whatever I used to get the images from the camera: either digiKam on my desktop or ZoomBrowser EX on my laptop.

shiroin
09-12-2005, 03:54 AM
If your scaled images don't seem as sharp as they should be, it's probably due to your software's scaling method.

As you can see here: http://www.dylanbeattie.net/magick/filters/result.html (of most relevance is the Blur = 1.0 column) there are actually a lot of ways to scale an image, some much better than others. A lot of software just does linear or bicubic scaling, which are actually not very good conpared to many other methods. Lanczos seems to work really well, and is getting more widely adoped, but far from universally.
interesting note on methods of image scalling. i will look into that in detial when i have time. thanks alot.

Sean
09-21-2005, 12:51 AM
Dump card using Photo Mechanic (http://www.camerabits.com/)
Rename parent folder to CCYYMMDD-CardID-ProjectID-Subject_Description (20050921-01-PIX-Photoshoot_Kelly)
Use Photo Mechanic Batch Rename to rename photos to CCYYMMDD-HHMMSS-ProjectID-Subject_Description-Source (20050921-125213-PIX-Photoshoot_Kelly-D)
Use Photo Mechanic to embed IPTC copyright metadata (Copyright, Name, URL, Location, Keywords, etc.)
Backup to CD-R or DVD-R depending on size of dump
Move parent folder to appropriate project directory on project disk (nightly backup)

Where the photos goes from there depends on the project...the embedded IPTC metadata helps me prove ownership if anyone ever snags the photo and removes my watermark; the IPTC info is still embedded in the photo even if it's been cropped.

staereo
10-25-2005, 10:36 AM
1) Go through raws, tuning exposure and dropping bad shots.
2) Transfer from card to hd.
3) Batch convert to TIFs.
4) Run each through Noise ninja with the 20d noise profile iso matched, then tuned.
5) Run each through photoshop 7, despeckling if need be, then sharpening with fred mirandas 20d sharpener plug in; or fred mirandas edge sharpener action when the iso was high. Then I adjust the levels of light/contrast/colors.
6) Delete all pics except RAW and final copy.
7) Burn to dvd and check the DVD.
8) Crop/resize/optimize/upload pics to be on website.
OR
8) Crop/resize/convert colors to print.

Bruce

Drngd Kreationz
10-28-2005, 10:41 AM
My workflow

Scan through photos in cam to weed out obviously bad photos
transfer to "latest upload" folder
Re scan through photos in D-Image study them to see if they said what I wanted. Take notes and look over possible improvements and variations to the shot if i were to do it again.
Batch rename photos and move into an archive folder (folder with matching name).
open photoshop and process each photo individually (clean up any noise, White Balance issues/color correction, rotate, crop, etc)
save under an "Edits" subfolder in archive
Batch process Resize to the Site upload folder
Apply Watermark to photo
Upload to site
Create Back up CD/DVD

Monthly back up of entire Archives

didjiman
11-03-2005, 04:12 AM
For me, resizing for the web I usually stick with 640x480 or 800x600 pictures.

However, since getting my DSLR, the aspect ratio is not exactly square, so I've started going 720x480 and 900x600.

That's what the cropping tools are for :-)

KaineMaxwell
05-11-2006, 11:46 PM
So 600x800 for vertical and 640x480 is a good standard to resize pictures to? I usually try to stick with that for web publishing.

Ami Yuy
05-12-2006, 02:17 AM
1) Download photos to HD into folders named for event and if it was more than one day, into the days
2) Organize my favorites into a "Best" folder (Sometimes even a "Best of Best" one when I have a TON or a lot that are similar)
3) Open up Photoshop CS2
4) Open similar photos together (same location/lighting/subject)
5) Crop if necessary
6) Save in "Edits" folder inside the "Best" and add "amiyuy-" to beginning of file name
7) Levels
- Clip ends if they're flat-lined
- Possibly adjust mid-tones if needed, but avoid if possible
8) Curves (where I do the most work for colors and such)
- Mini-S shape - try and make the tones and contrast nice
- Adjust the various RGB channels from the midpoint if I need to fix some coloring
9) Saturation +10 (sometimes bring it down or up, but generally 10 works perfect)
10) Touch-ups to eyes and skin if needed
- Duplicate background/main image layer
- Dodge eyes (10% and generally go over two or three times)
- Healing brush for dark circles under eyes or blemishes
11) Save again
12) Run my Cosplay resize & watermark Action that is most applicable (I have 8 different ones depending on how much work I did, their orientation, and the output size I want (640 or 600))
13) Save for Web - make it as close to 120k as possible, put in "Final" folder inside of "Edits"
14) Repeat

And then at some point or another upload them. ^^

And at some other point they eventually get backed-up. -_-;

natsukoarts
05-12-2006, 04:28 AM
I usually resize pictures to less that 640X480, but to 320X4xx.

Workflow depends on the picture, usage, lighting in the shot etc...

sometime chanells, sometimes levels, other times hue/saturation or brightness...

jtnishi
05-13-2006, 12:14 AM
Size is dependant on a few factors.

For cosplay.com galleries, you will ALWAYS do best if you resize the picture to 640 on the long side in your app. I find that cosplay.com's image resizer is a bit soft for my taste, so especially if you are concerned about keeping your edges sharp. This is probably true of any picture you're uploading through any gallery software where there's an upper dimensional limit, since algorithms in most gallery software tends to favor remaining relatively fast. For the pictures I put in my cosplay.com gallery, I resize to 640 on the long side (I especially like 640x640 for impact here), and sharpen last, to keep edges sharp, to give the image a lot of pop. I also lock file sizes down to 120KB.

Speaking of file size, as far as web display is concerned, the bigger concern instead of dimension should be file size. 640x480 or 800x600 should be less of a concern than making sure your images don't grow oversized and suck too much bandwidth or take too long to load. If your tool allows you to limit file sizes, you should aim to try to keep your pictures under a specific file size limit. I happen to like cosplay.com's 120KB limit, since my experience is that for an 800x600 (or 900x600, in the case of my camera) image, it produces a reasonable quality while not sacrificing too much in size. If your bandwidth is limited, go smaller.

Beyond those, within the constraints of monitor size and page design, best impact tends to be larger over smaller. I aim for around 600px on the short sides, myself. Those are my practices, and they tend to work well for my tastes. Your mileage may vary.

staereo
05-13-2006, 06:53 AM
Boy. Jason and I seem to always be on the exact same wavelength.

So I will just sum up my post to one word. 'Ditto.'

Bruce

WingedPower
05-16-2006, 01:11 PM
At end of day(at convention or after shoot/event):
- offload 1 copy to laptop/usb-hd
- offload 1 copy to dvd
- if raw images, generate quick previews.
- quick review of images on HD for "keep/no-keep" selections
- keepers are given basic treatment:
- if raw, convert to 16bit/channel tiff
- if JPEG, use as-is.
- standard meta tagging information(EXIF/IPTC) added
- awb/levels (need this less and less as I almost always re-custom-wb in-camera now)
- sharpening (via USM in PS. Normal sharpening creates too many artifacts)
- resize+watermark
(2 flows, depending on which computer I'm using. 1st flow uses PS's batch actions to do all of the above at once. 2nd flow uses ImageMagick tools scripted. Either one works just fine, just depends if I'm using my Mac or my Linux system.)
(resize is done to 800 max longest dimension. This is a good max size for most monitors. The scripting will take care of maintaining the ratio of the image. The online gallery software will take care of image resizing to smaller sizes.)
- If keywords/descriptions need to be added, they are added here upon final review.
- upload pics to webserver/gallery

On arrival at home, or access to fast internet to server at home:
- sync images/photos to main server's hard drive. (1 copy on usb hd, 1 copy on dvd, and 1 copy on server)

For RAW conversion, I use dcraw(command line utility) as it works on any RAW file. PS has raw support, but dcraw is more easily scripted.

deleriumx
05-20-2006, 12:31 AM
this thread would be really useful with photos that help explain what you are doing as you are doing it. I decided to give this a shot to see if i could somehow explain my workflow.. so here goes. let me know if it is helpful at all.

When going through my photos i noticed i tend to merge my layers a bunch so i decided to start an all new photo edit from an out take to show you each step. so the composition on this photo isnt the best, but you get the idea.

I want to preface this by saying that i do not usually do this kind of editing with regular hall cosplay pics. for those, i just do the first couple of steps only. This process applies to my more artistic photoshoots.

Step One: After sorting my photos into good/bad/maybe, i pick one to start working on. So this is the original photo straight out of the camera.
http://www.seventhstranger.net/edits/step1.jpg



Step Two: Crop the photo if needed
http://www.seventhstranger.net/edits/step2.jpg


Step Three: It turns out this photo needed no level adjustments. If it had needed it, i would have done that in this step. So instead, i went to my next step of fixing skin blemishes and undereye circles with the healing brush.

http://www.seventhstranger.net/edits/step3.jpg

Step Four: This step is optional, and i did it really quick just to show what i mean. Here i use the clone tool to remove stray hairs. Sometimes this can be a real pain and take a long time to get it to look right. I could spend ages on this step, in this photo, obviously, i did not. I only really do this if the stray hairs are really out of control and distracting.
http://www.seventhstranger.net/edits/step4.jpg


Step Five: This is subtle, so look closely! Here i removed the red from the eye using the brush tool set to Color mode at about 20%. Change the color to black and color over the red in the white of the eye. it will basically desaturate it. the veins are still there, just not red anymore. Later i will brighten them to make them nearly invisible. I also sometimes darken the lashes in this step. i did it a little bit here.

http://www.seventhstranger.net/edits/step5.jpg


Step Six: This is my favorite step! But it can go really wrong if i'm not careful. It's the dodge and burn step. Here, i brighten the eyes, darken the lashes, make shadows stand out and sometimes even lighten the background too. This is the step that makes the most difference in the photo, but as i said, it can also make it look really unnatural if used incorrectly.

http://www.seventhstranger.net/edits/step6.jpg


Step Seven: Also optional, sometimes i smooth the skin out a bit with the blur tool just to give it a softer look. I did it on the skin and hair here. and that makes the final photo!

http://www.seventhstranger.net/edits/step8.jpg


Okay so thats it in a nutshell. sometimes, i go a little crazy with my edits when i am feeling particularly artistic. They are basically the same steps i listed above, just in greater quantities. Here's an example of some overboard editing:

Before (http://www.seventhstranger.net/edits/wow1.jpg)
After (http://www.seventhstranger.net/edits/wow2.jpg)

thats all!

Ashurachan
05-22-2006, 04:42 PM
- copy everything on hard drive
- delete from CF card
- import everything in library software (previously : iPhoto, then : iView Media Pro) and tag with con name
- quickly view everything full screen
- despair because everything's just crap
- forget about taking these uglies for about six months (optional, but frequent)
- browse through everything again
- when not too bad picture found :
* open in Photoshop
* play with levels, exposure, whatever to make it look more acceptable
* if needed, play with colors - especially saturation.
* remove red eyes
* if it's a portrait and needed and worth the effort, smooth skin (I don't have just one technique, but several that complement each other, and it would be pretty long to explain)
* automatic contrast (I don't trust my screen for this)
* if actually really like picture at this stage, save for web, jpeg, 80 quality, baseline (for future prints)
* if composition OK, resize to 500*750 ; if not, crop to 500*750
* paste watermark and place it the corner where it distracts the less from general composition
* save for web, jpeg, 60 quality, progressive
- rince and repeat
- after last picture, feed folder to thumbnail-making program (*)
- open resulting page in GoLive
- copy/paste bits of code from already done page, get layout ready
- make con icon (small psd ready to have its text changed -> gif)
- write short comment about con
- order pictures in page (<- reason why I do not use some premade php tool for my galleries)
- if bored, add comments under pictures
- create link to page on website homepage
- upload everything new/changed with ftp
- change MSN name to reflect site update
- tell friends on MSN I updated
- post best pictures on LJ
- post announcement on cosplayforum.com
- go to sleep 'cos it's 3AM and I have to get up at 7 for work.
- forget to save on CD-R or DVD-R
- get all catastrophy when hard drive crashes (<- happened last month)

(*) if someone knows a simple one for Windows (no need for fancy pages, just creates thumbnails, and a page with links to full pictures), I'd be glad to hear its name.

KaineMaxwell
06-23-2006, 11:53 PM
For me:

1) Download pics off camera to folder marked whatever the subject is.
2) Use ACDSee Classic to look through them and weed out bad ones.
3) Open Paintshop Pro.
4) Go through each one with One Step Photo Fix then rotate and resize to 600x800 or 600x450 (or whatever its auto guide sets it too).
5) Touch up any with Sharpening, cropping, etc.
6) Save.
7) Upload somewhere.
8) Sleep.

I should really keep the larger 1600x1200's somewhere. i'm a bad photographer..

Zetchi
06-24-2006, 01:09 AM
1. Take out my memory stick from my camera and stick it into my USB connector to take the pictures off the memory stick.
2. Move the files into a folder (usually under events >> anime conventions >> and whatever the name of the convention was, along with the dates
3. After all the files are moved, I check to see which ones need work (poor lighting for example)
4. Open up Photoshop and usually click on "Auto all levels". If it doesn't fix the picture, I tweak it myself using the brightness and contrast options.
5. Once that's done, I upload them and then link them.

I usually keep my size at 600x800. The quality works for me and I usually come out with good pictures, well I can say that when I take the pictures.