If anyone can clue me in as to where/how to begin to merge multiple sequence RAW or JPEG files of say someone skiing, into a single frame, I would really appretiate it. I have these sorts of shots and I want to combine multiple seqencial frames into one new image. What sort of software/basic technique is needed? I have DPP and photoshop 7.0 and shots have been taken with a 1D Mark II. Many thanks in advance-

Do you mean like this:
Matt Philbin wrote:
Nice one ^^. I did my best from Chicago, but it's quite cold here. I put this together after I got in... 70-200 + 1.4x...

It would be easier with Photoshop CS3. Here is how how I would go about it with PS7. Open all the images in Photoshop. Zoom out so you can more or less line up all the images in the Photoshop window. This should give you a good Idea of how many frames wide the composite will be. Then open a new document twice as high as a single frame and wide enough to accommodate the composite with. The drag and drop all the images on to the new document. They will each be in it own layer. Now you want to align these layers. The skier and you camera were moving as you captured these images. The background was not. So your going to line up the image backgrounds. First do a quick rough arrangement of the layers. Next the easy way align the best is work with two layers at a time. Arrange the layer stack so the left most image in on the bottom. The image to the right of in the next layer up and so on to you reach the top. Turn off the visibility of all the layers but the bottom two. Next set the blending mode of the layer on top of the bottom image to difference. Move the layer about a bit till the overlapping area is as black as you can get it. This will align the layer the best that they can be. Next turn the visibility of the bottom layer off. The bending mode of the layer on top of it back to normal. The the layer of top of it visibility on and its blending mode to difference. Align the layer. Repeat till all layers are aligned. Then turn all the layers visibility on and the top layer bending mode back to normal. Then add a reveal all layer mask to all the image layers except the bottom image layer. Next paint wit black on the layer mask to hide the part of the layer the may be hiding the skier in the layer below it. Start with the bottom layer mask and work your way to the top.

It would be much easer with CS3 it has some new features. First it comes with a script "Load images into a stack". You can the select all the image layers and use the new auto align layers features. You could follow that with the new feature auto blend layers. At that point your done.

Mr Mouse wrote:
It would be easier with Photoshop CS3. Here is how how I would go about it with PS7. Open all the images in Photoshop. Zoom out so you can more or less line up all the images in the Photoshop window. This should give you a good Idea of how many frames wide the composite will be. Then open a new document twice as high as a single frame and wide enough to accommodate the composite with. The drag and drop all the images on to the new document. They will each be in it own layer. Now you want to align these layers. The skier and you camera were moving as you captured these images. The background was not. So your going to line up the image backgrounds. First do a quick rough arrangement of the layers. Next the easy way align the best is work with two layers at a time. Arrange the layer stack so the left most image in on the bottom. The image to the right of in the next layer up and so on to you reach the top. Turn off the visibility of all the layers but the bottom two. Next set the blending mode of the layer on top of the bottom image to difference. Move the layer about a bit till the overlapping area is as black as you can get it. This will align the layer the best that they can be. Next turn the visibility of the bottom layer off. The bending mode of the layer on top of it back to normal. The the layer of top of it visibility on and its blending mode to difference. Align the layer. Repeat till all layers are aligned. Then turn all the layers visibility on and the top layer bending mode back to normal. Then add a reveal all layer mask to all the image layers except the bottom image layer. Next paint wit black on the layer mask to hide the part of the layer the may be hiding the skier in the layer below it. Start with the bottom layer mask and work your way to the top.

It would be much easer with CS3 it has some new features. First it comes with a script "Load images into a stack". You can the select all the image layers and use the new auto align layers features. You could follow that with the new feature auto blend layers. At that point your done.

Haha. CS3 might be worth if you wanted to only do this one thing. A huge paragraph worth of instructions for CS7 vs. 3 sentences for CS3. haha

Actually for him CS3 is worth the upgrade because he shoots RAW. Adobe did sell a RAW plug-in for PS7 but is is crude compared the ACR 4.4.1 available for CS3. The improved RAW work flow you get with the Adobe Bridge/CS3 is quite mature compared to any PS7 raw work flow.

jdryan3 - yes, more or less, thanks.
mr mouse - thanks for the tutorial, I don't have CS3 because I just stepped it up to a 1D body and I am still learning and figuring out which PP programs suit my needs. I will give your method a try later this weekend, and I really appretiate the help. Cheers-

zlatham wrote:
If anyone can clue me in as to where/how to begin to merge multiple sequence RAW or JPEG files of say someone skiing, into a single frame, I would really appretiate it. I have these sorts of shots and I want to combine multiple seqencial frames into one new image. What sort of software/basic technique is needed? I have DPP and photoshop 7.0 and shots have been taken with a 1D Mark II. Many thanks in advance-

If you want to use photoshop for it, you can follow the advice others have given. An alternative workflow would be using hugin (or similar) and having it match up all the pictures for you. You can then have it output the images as a layered tiff (or photoshop file) and you can do your masking with photoshop/gimp/paintshop.

As far as the masking goes - I would think it easier to choose one frame to use as the base, and keep only the subject from the rest of the frames (no background).

invalid2 wrote:
As far as the masking goes - I would think it easier to choose one frame to use as the base, and keep only the subject from the rest of the frames (no background).

If the images were taken with a stationary camera yes. If the image were take while panning with the camera they need to be stitched like a panorama.