Turn 2D images into Anaglyph 3D using Photoshop

Showcasing images and photos is one thing, but really capturing people's attention in a media-soaked world is another. That's why 3D has recently become so popular again. There are two main types : stereoscopic, where two cameras are used to capture an image, generally to create a 3D movies; and anaglyph 3D.

Anaglyphs are an older technology and rely on color channels, but they can be used in media such as print. Best of all, you don't need to have created a 3D image to start with. Using Photoshop, you can manipulate the color channels to create various types of anaglyph 3D from an existing 2D image. For this tutorial, you'll need a pair of green and magenta glasses to view the images. The other type have red and cyan lenses.

Step 1 : Simple manipulation

The essence of creating anaglyph 3D pictures from a single image is to manipulate the color channels. Start by loading an imago, and in the Layers palette click on Channels and then the Green channel. The image should turn grey when only this channel is selected.


Step 2 : Instant depth effect

Click on the Move tool in the Tools palette, and click and hold anywhere on the image. Drag about 3mm to the left and let go. Now click back on the RGB channel in the Layers palette to recombine the image. Put on the 3D glasses and you should sec that the image is behind the monitor front.

Step 3 : More depth options 

 
This next effect requires an image where the content is angled across, so that objects on one side are near the camera and those on the other side are further away. This is how to enhance the effect with 3D. Go to the Green channel again and then press Ctrl-A to select all.

Step 4 : Sliding depth 

Go to Edit - Transform - Scale and grab hold of the control point halfway down the left side. Click, hold and drag to the left to scale the Green channel to the left. Now when you recombine the channels, the right side of the image will appear at the front of the monitor while the left will recede into the distance.


Step 5 : Front to back 

Select the Green channel and press Ctrl-A. Go to Edit - Transform - Distort. Grab the top-left control point and move it left. Repeat with the one on the top right. Combine the channels and when you look at this one. the bottom of the image appears on the screen, then recedes as you look up the to the top.

Step 6 : Two-layered system 

 
Let's look at some other things that can be achieved by splitting the image into two depth levels. Load a portrait photo and toggle Quick Mask to On. Click on the Paintbrush and set the opacity to 100%, the foreground color to black and the brush hardness to 75%, Paint over the background, leaving the figure.

Step 7 : Cut out the figure 

Zoom in to go around the edge of the figure and use a small brush to paint around any strands of loose hair. Once complete, toggle the Quick Mask off so it becomes a selection. Then, press Ctrl-C to copy the figure. Select the Layers palette and create a blank layer Press Ctrl-V to paste the figure in.


Step 8 : Fake some background 

Moving the green channel will cause a large streak to appear. Turn off the visibility for Layer 1 and select the Background layer. Zoom in to 100% and click on the Clone Stamp tool. Set the opacity to 100% with a blend mode of Normal. Clone scenery from the left of the face, over the left side of the face.

Step 9 : Move the channels 

 
Click on the Channels lab in the Layers palette and select the Green channel. Click on the Move tool and move the channel to the left as demonstrated earlier. Click back onto RGB. then turn the visibility for Layer 1 back on. Press Ctrl-Shift-E to flatten the layers, then crop the magenta strip on the right.

Step 10 : Graduated backgrounds 

Next, we're going to create an image with a separated figure, but a background that has a graduated aspect for the depth. Turn the Quick Mask mode on and select the Gradient fill tool. Click and drag from halfway down to the bottom of the screen. The top half should be red.

Step 11 : Remove the figure 

 
Select the Paintbrush and set the hardness to 75%. the foreground color to white and the opacity to 100%. Zoom in and use a largo brush to remove the rod mask from over the figure and, in our case, the fence, leaving the edges. Use short strokes that are easy to undo if you make a mistake.

Step 12 : Close up work 

 
Reduce the brush size for removing the mask by the edges. Where the hair is fuzzy, reduce the brush hardness to 25%, then increase it again. Mistakes in tho area where there is 100% red can simply be painted back over, ones where it is graduated have to be undone and reapplied.

Step 13 : Apply new layer

 
If you can't see the edges of the figure.toggle it on and off for a better view. Also do this to make sure you haven't missed any areas. Toggle tho Quick Mask off to reveal the selection. Press Ctrl-C to copy the figure. Create a new blank layer in tho Layers palette and paste it in with Ctrl-V.

Step 14 : Tidy up around the figure 

 
Turn Layer 1's visibility off and select the Background again. You now need to clone in over the left side of what was copied for Layer 1. This means the left side of the girl and branch and the under side of the fence. Use the Clone Stamp tool with a 100% opacity and 0% brush hardness.

Step 15 : Move into position

When cloning, extend the scenery without repeating textures and shapes. Next, click on the Channels tab and then the Green channel. Select the Move tool and move it to the left. Click on RGB to recombine and then toggle Layer 1 back on. Merge the layers and crop the band on the right to finish. 

Read More

How to : Create Awesome HDR Photographs

Most of you may be now familiar with HDR or High Dynamic Range in games. HDR enabled games look amazingly lifelike with better lighting and shadows and the clouds look ultra-realistic. What actually makes HDR images different from traditional images is the fact that they have a higher range of intensity levels (this article is meant for a layman and we are therefore not using technically correct words like Radiance and Luminance), and it is more like the way we perceive through our eyes.

Digital cameras are cheap these days, and you get image quality that rivals that of SLRs of yesteryear. But if you observe closely, there is still a lack of realism in the images that are obtained by these digicams. The problem is that the CCD is simply unable to capture the complete range of intensity of light in an image at one shot. An example of this is that the shadows are not dark enough, highlights are not bright enough or at times are too bright, and then there is the loss of perception of color of the light source. This is where HDR photography comes into play.

In HDR photography it is possible to capture and display the complete range of intensity of light and color that makes the photographs look lifelike with a richer contrast that doesn't look artificial (usually). Contrary to popular perception, HDR photography does not require an expensive digicam. Even entry-level cameras, which allow you to set different exposure levels or shutter speeds, can do a good job. All you need to do is take photographs of a scene with different exposure settings. If your camera supports Exposure Bracketing, then select the Continuous Shooting mode with the camera set to Aperture Priority and select an exposure increment of +/-2. The camera will automatically vary the shutter speed each time the shutter button is pressed and vary the exposure. The larger the number of photographs and the larger the exposure difference, the better will be the results you get.

The most common technique for HDR photography is to shoot a number of photographs of a scene using different exposure settings. Thus, some photographs are underexposed, while some are overexposed. The underexposed photographs capture the details in the lighter and reflective areas better, while the overexposed photographs capture details in the darker areas. These images are now combined into a single image, taking the better things from each image, thus resulting in an image that contains a larger range of light intensity. This method is known as Exposure Blending and there are several programs available that can do it.

Since HDR images contain a huge amount of data and it is not possible to display it as it should actually appear on conventional CRTs and LCD monitors, a method known as Tone Mapping is employed. Tone Mapping attempts to compress the entire dynamic range of light intensity into a viable range that can be displayed on the above display devices. This produces an image that retains realistic color and contrast.

We shall now explain how you can actually go about getting an HDR image using a free program called Qtpfsgui (and optionally also using GIMP or Photoshop). Qtpfsgui can be downloaded from http://qtpfsgui.sourceforge.net/download.php. Do not forget to download the DLL package also.

You need to have at least three photographs of the same subject with different exposure: underexposed, normal exposure and overexposed. You can achieve this by either varying the EV settings on your camera by setting it to EV-2 (underexposed), EV 0 (normal exposure) and EV+2 (overexposed). Or if you have a camera that allows you to set shutter speed, you can set the shutter timing to 1/500s for underexposed, 1/25 second for normal exposure and 5 second for overexposed.

If you are working with RAW images, it is a good idea to use noise reduction facility in your image editing software for each of the images before proceeding to the next step. This is because the tone-mapping procedure introduces some degree of noise in the resultant image and if the noise is reduced at the start, then it will be lower in the final image as well. With the photo set ready, you can now proceed to create an HDR image using the software.

1. Download and install Qtpfsgui and also extract the DLL package to the same folder where you installed the program.

2. First click on the New HDR button. In the next window that pops up, click on Load Images button and choose the three photos you prepared.

 

 Qtpfsgui will try to automatically determine the exposure value from the EXIF metadata in the image, but if it fails, then you will get an error message and will have to specify it manually.

Select the Image Filename on the left and set the EV value for individual images using the slider in the right hand side of the window.

 3. Select the Auto Align Images checkbox so that the program compensates for or corrects any misalignment in the consecutive shots. Let it be set to "hugin's align_image_stack". Click on Next and skip the next window, which is the Editing Tools window, as it requires some practice to modify anything here.

 4. Skip the next window too, where there are a number of predefined profiles to choose from, which give you slightly different results. Let default option remain chosen here and click Finish. The next window will show you the raw HDR image, but this is not our final result. This still requires tone-mapping so that it can be displayed correctly in a low dynamic range image format (jpg).

 6. Now click on Tone-map HDR button in the toolbar. A window will pop up with a number of tone mapping algorithms and parameters.

7. Now choose Fattal tone mapping operator and increase the Noise Reduction slider to a value between 0.05 and 0.5 and reduce the Beta slider a bit. Click on the green button Apply. You can try changing the values a little to get better results. 

 

 You can fine-tune the result by clicking Adjust Levels button. After you are satisfied with the results, save your image file by clicking on Save as button at the top. Make sure to choose the Result size as the same as the size of the original image while saving the final image.

 8. Now choose the Drago tone mapping operator tab and repeat the procedure in the previous step. You should finally have two tone-mapped JPG images.

9. Open both the tone-mapped images in GIMP or Photoshop. Use the image tone-mapped with Drago for the first layer and put the Fattal tone-mapped image as the second layer. Set the Layer mode to Overlay and adjust the Opacity such that it looks good. Or you can go the other way round and use the Fattal image for the first layer with the Drago image as the second in Overlay mode and set the Opacity to around 50%.

There is yet another way to get nice HDR photos - by combining Fattal and Mantiuk techniques instead of Fattal and Drago. When using the Mantiuk option, you should adjust the pre-gamma slider to around 0.5 for a slightly lighter image and increase the saturation slider to around 1.0.  The tone-mapping process remains the same as before. You can open Fattal image as the first layer and Mantiuk as the second with Layer mode set to Soft light and Opacity adjusted to give a good look. For a dramatic looking (but seemingly artificial) output, you can apply a slight blur mask to the top layer.

 

Summary

10. Lastly, you can post-process the final HDR output image in Photoshop or GIMP to give it a more professional look. You can increase the contrast by applying Curves or Levels, adjust the color or white balance, etc. Sometimes you get strange ghosting artifacts in the final HDR photos, which can be fixed using the original image with Layer mask.

Creating an HDR photo (albeit a fake one) from a single JPG or RAW file

For those without a decent digital camera, all is not lost. You can still create a fake HDR out of a single photo. Open the RAW image file in Qtpfsgui using File -> Load HDR and then tone-map the file directly. Though there's little sense, you can try to achieve fake HDR look with single jpg or raw file. You can directly load a raw file in Qtpfsgui via the File->Load HDR and tone-map directly single raw file.

 

 For creating overexposed image

 For creating underexposed image

Another way is to create two different "exposure" versions of the photograph using the Curves function as shown above. Click in the middle of the curves graph and drag it to the bottom to create an underexposed version and save it as underexposed.jpg. Similarly, drag up to create an overexposed version and save it as overexposed.jpg.

We hope that you will start creating your own HDR photographs using our guide. Please use the space below to let us know about your results.

Read More

Try a Body Tattoo Virtually on Your Photo

Have you ever wondered what you’d look like with a tattoo? With Tatmash you can now easily add a tattoo to your photo and see what you’d look like without a tattoo needle even pricking your skin.

TatMash : Try On Any Virtual Tattoo Design on your Body

Using the service is very simple and free. You can select from their collection of tattoo designs that you want to try out. If you don’t see anything that you like, you can also upload your own design. 

Just make sure your design has a white background and Tatmash will automatically handle the rest so your design to come out great.TatMash is a unique web app that lets you preview different tattoos and see how they look on you. This way, you will know what specific tattoosuites your body.

Read More