Welcome to WebmasterWorld Guest from 188.8.131.52
Forum Moderators: not2easy
There have been a couple of excellent threads here on processing photos for web sites, but there is a limit to how much *magic* can be performed if the original photos are not properly taken - making it much more difficult to edit and alter them to be properly presented on a web site -more work for the web designer, and potentially more costly for the client.
The question: How do you take the original photographs to present products on a web site for maximum quality and presentation?
1) Lighting issues (they're usually too dark)
2) Background (significant when trying to make a "selecton" to adapt to the chosen background image/color)
3) Scanning issues - size, resolution, etc.
4) Color issues - the photos sent often do not reprsent the actual color of the products
There are more issues concerning taking photos, but some basic priciples and precepts would be greatly appreciated and helpful.
#1 you must have good Photo edit software. Photoshop or PaintShop Pro is good as well. Paintshop I think has a free download.
With this software you can brighten the picture and Sharpen the image until it is pretty good.
>>2) Background (significant when trying to make a "selecton" to adapt to the chosen background image/color)
Not sure what you want to do here. It is the background of the site or do you want to cut out the background of the picture so that it has a Tran. background????
>>3) Scanning issues - size, resolution, etc.
I always like to scan all images as large as possible. with the largest RES that i can get. Files get rather larger doing this and the scan can tie up your machine for awhile BUT you get the cleanest image and you can resize a compress to make the best load time with a great image for your surfers.
>>4) Color issues - the photos sent often do not reprsent the actual color of the products.
Again using one of the above softwares that I listed at the begining of this post can change the colours etc.
Marcia any else i can help with just LMK
My opinion is that a professional product photographer is worth every penny. I have clients that range from using professionals on one end (our company offers the services of a top notch pro) to disposable camera photos from the pharmacy. Sometimes the disposables work, but it's usually a crapshoot.
One of the keys is the angle of view. When the item is on screen, it looks more natural to the surfer if it was shot head on. When you shoot head on, the lighting needs a professional eye to get good results. A single flash usually is unsatisfactory.
If you are going to make a selection in your image editing software, a uniform background is a big help, although it's not essential. Photoshop can do wonders if you blow up the image to 400% or better and use the lasso tool that lets you click from point to point to define your selection. In fact, let me recommend an incredible book about Photoshop. It's called: Real World Photoshop 5: Industrial Strength Production Techniques.
I've devoured 8 or 9 books on Photoshop. Most of them are about creating eye candy. The thing that makes this book different is that it zeros in on the things you really need to know in a production environment. A lot of it is for pre-press, and not web, but a lot of the tricks they offer you'll never learn any other way.
It's not really necessary to scan any larger than 2 times the resolution you will need in your final image. They key is that the information must be in the original photo. Use the right camera and lens to get an image that fills the print. If you are doing homegrown shots but can get oversized prints, that's a big help.
Here's another method that gets overlooked. If the product is relatively flat, scan the physical object itself and cut out the middle man -- the camera! I've had some great results with needlepoint canvases, for example.
If the image is too dark, or needs a lot of color correction, the first port in the storm is the Photoshop "autolevels" treatment. It's amazing what that alone can do.
Enough already. Hope this helps.
As for manipulating the images, I've been using Lview since it was free. I upgraded to LviewPro [lview.com] a couple of years ago. It's weak on text effects, but cuts to the chase on the majority of tasks I've had to tackle (usually lightening or sharpening).
Also, if you aren't about to layout some heavy cash for Photoshop you can always take a look at PhotoImpact from Ulead. It's only about $100 and it absolutely rocks. It can also use 99% of Photoshop plugins. It's all I use.
That said, we all have our favorite image processing and graphics tools. The key is finding one that you are comfortable using and gives you the results you want. For example - I've never been able to get a good handle on Photoshop but Photoimpact was an easy pickup for me. I have friends that are the other way around and hopelessly addicted to Photoshope like tedster ;)
1. Color Balance - neutral (gray) tones should have an equal value of rgb example:(67,67,67)
2. Highlight Density - highlights (lightest area) should have a rgb value of 240,240,240 only catch lights (small areas, [ 8 pixels x 8 pixels] such as a bright reflection) should have a value of 255,255,255.
3. Shadow Density - only the darkest areas (8 pixels x 8 pixels) should have a rgb value of 5,5,5 others should be at 20,20,20.
4. Detail - the mood of the photo should set the sharpness of the photo.
5. When all possible, have the photographer, take a test shot.
This test shot should be a 8 x 10 color chart, that has wide variety of hues, including a range of neutral whites to grays.
This shot should be take directly after the product shot using the same lighting conditions and camera settings.
Rather than having the product shipped to you for color matching, recieve and match the chart.
This is a proven method used in the printing and publishing industries for many years.
The only difference is printing works wth cmyk in unequal balances.
You also reminded me of a "secret" that I might as well take out of the bag.
If the image you are preparing for the web has ever been through the print process, it has been through CMYK color space. CMYK has a limited gamut, compared to RGB, which as you know is the color space monitors use. Your image has been "dulled down" to the more limited CMYK color gamut.
Such images have a lot less "pop" than they could. I say they have a print world ball and chain around their ankle. This is the web, and we don't need to pay that "tax" on our images.
I ALWAYS test an image for color enhancement by boosting the color saturation. Only occasionally are the results no good, perhaps rather garish -- but that's very seldom. Usually, the extra gamut available in RGB responds beautifully. Your website's customers will subliminally feel that the products are more appealing. This takes less than a minute to do, and it can mean higher conversion rates, better profits, happier clients.
I have PSP6 (which I've never found easy, as they claim), and Corel Photopaint, which I haven't had time to delve into yet, but it's based on "object" logic like another program I've used, which I love.
I use Photoshop 5 almost exclusively. Now I can that there are tools and capabilities in Photoshop that I haven't even discovered yet. Time to dig out all the books again, or maybe get a new one.
(Academic pricing on software is the greatest!)
Yes, most importantly, I want to be able to separate the item from the background, which is virtually impossible because of the closeness in color and value - all the photos seem to be dulled out, with little contrast. Not having seen the original objects, I use my best guess for what the colors should be, which generally works - except for the background, and also, getting enough clarity.
I'll try working with the shadow and highlight density, and the sharpening in stages while reducing size - these photos come in at about twice monitor screen size, very fuzzy.
These are flat rate sites, so I can't spend an hour on each photo. If need be, I'll have the items sent to me, and do the photos myself. Not a bad idea, a new camera would be a nice *toy* to get, and this would provide an excuse. I have one of those 500 watt lamps, so that could help the lighting situation if I do the photos.
These people will not pay for professional photography, I have to wheedle them into paid hosting with a package deal for hosting/maintenance/search engine work on a continuing basis, thanks to a couple of "business" type threads here, and this is working - not one has turned the deal down, and it gives me plenty of leeway for results and changes.
Thanks a million, I will be back for more specifics!
You could try going to L-a-b color space and making a copy of whichever channel shows the strongest edge. Then pop the contrast of your copied channel to the point where the edge is dramatic. Sometimes you can also get a good edge with Threshhold.
At that point the magic wand will probably make a selection on your copied channel for you with no problem. Then just switch back to regular "layers" view and you've got it.
When this works, it's maybe a 3-5 minute operation, max. Problems sneak in with wispy, feathery edges which the wand wont select properly, but even then you can sometimtes grab it with Color Range. It just takes a little more trial and error, and maybe some spot cleanup of the selection afterwards.
Or, you could buy a new camera and enjoy!
you may want to consider having your client make a photoCD. it isn't expensive at all. thus, you will receive the photos in digital format. if you are receiving a print from your client, ouch. the print has lost a huge amount of shadows, tones and highlighted detail before you even start. very difficult to work with. i know of an excellent kodak award winning company that takes orders on-line. since i'm not suppose to list urls, if your interested email me.
Many (most?) film developers offer photoCD's, so this can help a lot.
Here's another option that one of my creative clients came up with in a pinch situation -- I just finished processing a bunch of images that were color photocopies of the product, which was essentially flat.
I was a little dubious at first, but it was a weird situation, so we tried it. They were a breeze to scan and reduce. I probably got them done twice as fast as working from either prints or digital images. The size of the photocopy made the difference, I believe.
>My opinion is that a professional product photographer is worth every penny.
I wholeheartedly agree. But sometimes you can't afford that. So, if you can't afford that, then some things to keep in mind:
This is a biggee. Ideally, you want soft light, unless you are going for a drastic look. (Sunsets, for example) Making this happen inside with flashes is difficult, unless you are a pro. And with flashes, you get the red eye problem which makes "People of the damned" type photographs. A simple solution to the problem is, if possible, to take the photos outside, preferably on an overcast day. If you don't have overcast, early morning or late evening. And if you can't do early morning or late evening, I have seen photos taken in solid shade work well. (If you do this, be aware of your background and make sure you it is not brighter than what is in the foreground.) Avoid noon like the plague. It puts wierd shadows under the subjects eyes.
2) Some things to keep in mind when taking "People" photos:
--a)The height of you and the individual. Keep the camara the same level as the person you are photographing. An example: I am 5 feet 2 and 1/2 inches tall. Whenever my 6 foot husband takes a photo of me, I look about 3 feet tall. He is taller and angles the camara DOWN at me to get the photo. When I take a photo of him, he looks unusually tall because I have to angle the camara UP to him. (History buffs: see Leni Reifenstal, Hitler's photographer. She always pointed the camara UP to make him look taller than he really was.)
--b)The eye is the most important. Focus on the eye of the individual. It is the first thing most people look at. Particularly true with profile shots
--c)Sports shots: always include the ball, or object being played with.
--d)Make sure there is nothing distracting in the background. Keep the background simple, solid is the best. Some people take photos and don't keep in mind the background and they will wonder why there appears to be a tree branch growing out of their head.
--e)When you are taking a profile shot of someone, or they appear to be looking at something, make sure there is a little space in front of their eyes. If you cut it off abrubtly, the person looking at the photo will wonder, in their mind, "What are they looking at?"
--f)When you put a person shot on a web page, make sure that the person is looking "in" to the page. If the person is looking, or appears to be turned a little to the right, for example, put the photo to the left side of the screen. It will draw the viewer into the rest of the page.
--g)GET CLOSE TO YOUR SUBJECT. That is difficult with people because they want their personal space. Big lens is first choice, but can distort. An easy way to get around, and what I always do, is with the very first photo, stick your camara in their face. I mean *right* in their face. Six inches from their face. They will of course, lean back. But, when you pull back to about 3 or 4 feet, they will relax. You are very close to them, but they don't feel intruded upon...you got them over that when you put the camara in their face.
3)FILL THE FRAME! With product shots, there should be one thing to focus on. Let me give you an example. I did a quarter horse site for a client. I needed a photo of a particular mare. She sent me a beautiful photo of a barn with a beautiful scenic background. I called her back to tell her that she sent the wrong photo. She told me no, that was the correct photo, the horse was standing next to the barn. And she was right. If I looked carefully, I could see a tiny little horse standing next to the barn. It was about 1/16 of the entire photo. I didn't want a photo of the barn. I wanted a photo of the horse.
An easy way to do this is to take an 8 x 10 photo frame and "look at what you are going to photograph" first. Study the background and move close to the subject to make sure that the only thing in the frame is what you want.
4)Make sure you have a photo of the entire thing you are photographing. Another horse example: my client has sent me many photos of her horses, but will cut the bottom of the photo off at the horse's knees. That is not an entire horse, and it looks wierd to the viewer. (I have asked her if her horses "float". Of course, she has missed the point and said "Yes, they float, they are the finest ride money can buy." *sigh*)
5)Take a rectangular piece of paper. Draw lines on the thirds horizontally and vertically. You will find 4 points where these lines meet. These are your "action" areas. A dead centered photo is not pleasant to look at. Put the main "action" or "thing" into these four action areas. This is very true with a profile shot. The eye of the profile should be in one of these action areas. It just looks better.
6) Crop Crop Crop. Many people do not fill their frame, so crop out everything you absolutely don't need. I crop my photos very tight. And, it makes the photograph faster to download if you are not forcing unneeded pixels down the pipe.
These are just a few things off the top of my head. I could lecture all day on this, but I have prob killed this thread with this post. I will stop here.
true, many places can make a photoCD but is it in a near dust free environment.
lol to #4 by grnidone.
how does one explain to a client how to take photos without making it sound too complicated or insulting. i think i may have lost a client by getting too complicated on this topic once because i'm very anal about it. i sent examples of bad photos i found within seconds off the net, with explanations and everything, then never heard back. i don't know if this is the reason but since it was the last correspondence, not even a thank you for a reply, and this person was very keen on my work before this, never blinked at my price, and they contacted me. (i don't advertise because i'm not ready to do this, and maybe i will never be ready or willing.)anyway; therefore, i assume it is the reason. if i would have lived near this client i would have offered to take the photos myself but not possible. besides, no doubt it would have been less expensive to hire a professional photographer and studio in their area than to have me fly in and fumble around for a few days.
imagine 3d scanners like studio lamps available at your local home computing store, now that's a cool idea.
I have one client who is a very enterprising "do it yourself" type. In the beginning, the photos they sent were quite rugged to work with.
I just sent back the photos that were sub-standard with a simple note about why. No big explanations, just "too dark", "image too small" etc. Over 6 months of site updates, the photos got better and better as the client climbed the learning curve.
The payoff for their learning is that the site updates happen on schedule.
mousemoves, people will try to do that!! You might find this thread, related to that very topic, interesting to read: