Tag Archives: photoshop

Stitching shots – a Gigapixel panorama

So you want to shoot a dramatic landscape that requires an ultrawide lens. If you don’t have an ultrawide or you need more detail than your camera can capture you can always use photoshop to stitch more than one image together.

2000 pixels wide
2000 pixels wide

This image was composed of 9 separate frames shot with a Nikon 70-200 VR lens. I used the 70mm end as it has minimal distortion and hand held a series of shots starting from the left and panning right. It’s important to overlap your shots as this will make it easier for photoshop to find anchor points to compose your panorama. If you haven’t corrected your shots before merging you may find the horizon uneven. This is tricky to correct once merged so be careful when shooting to place your horizon in the same part of each frame.

When capturing your shots ready to stitch make sure that you set your camera to manual mode. Find an exposure value, white balance and focal length that works for the whole scene and lock them so you have no variation in your final image. Once you have your images ready to go you can load them using one of two methods.

  1. In Lightroom, Select your images then right click ‘Edit it > Merge to panorama in Photoshop.
  2. In Photoshop, click File > Automate > Photomerge.

Photomerge offers several options, by all means try them out but ‘Auto’ tends to work well for most situations.

Photoshop will import your images and attempt to arrange them by finding key points and creating layer masks to make a starting point for your panorama. Once the layers are complete you can use the brush tool to reveal or hide details from each layer.

The full size version of this shot is a rather frightening 1.24 Gigapixel image, larger than any DSLR currently available can produce. At 300DPI this image could be printed at over 2m wide.

This is just an outlined version of the process. For detailed workshops on capturing and creating with photoshop see the tutorials section for one to one sessions.

Detail crop from centre
Detail crop from centre

Tutorial – Tilt shift miniature effect with photoshop CS6

The tilt and shift effect is one of my favourite photographic techniques. It allows us to recreate the focus effect created by using macro lenses to photograph small objects close up, fooling the viewer into thinking they are looking at a model or toy town rather than real buildings and people.

First what is tilt and shift?

A regular lens is designed to sit straight in line with your sensor, giving a flat image plane and even focus across the image. Focus can be moved away from or toward the lens but stays the same across the sensor. By tilting the lens up, down, left or right we can change the focus distance across the image and selectively focus on one part of the picture.

In contrast, shifting a lens from the centre of the image plane changes the way your camera sees converging lines. When you photograph a building from ground level the vertical lines converge towards the top of the image. By shifting the lens upwards from centre vertical lines can be corrected giving a more realistic representation. This is often used by architectural photographers but with the huge cost of tilt and shift lenses this is not an option for everyone. Photoshop offers very good distortion correction so this is no longer a real issue on digital. We’ll be concentrating on the ’tilt’ effect.

When you take a picture of a city scene, even with a large aperture setting you will find that most of your image is in fairly crisp focus. When photographing on a macro scale the opposite is true, achieving a large depth of field is very difficult so we associate that shallow focus with photographing small objects.

This effect is in vogue right now and has been used a lot by the BBC and in various sports coverage. The recent series ‘Sherlock’ used tilt lens filming in its intros to great effect.

Fortunately for us Adobes photoshop CS6 includes some great new tools for creative focus effects. We’re going to be looking at the tilt shift and iris blur tools.

First we need an appropriate picture. I’m going to be using this shot of Liverpool One shopping centre.

Unedited capture pre tilt.
Unedited capture pre tilt.

I deliberately found a high vantage point and included lots of people in the frame. Buildings tend to convert well so I’ve made sure there are some in the top of the image. I used a 50mm lens on FX to keep a natural perspective. Wide lenses don’t tend to work as well so 50-85mm is a good place to start.

It’s easy to imagine that this image will work but it’s way too flat and has fairly even focus across the frame. First thing is to make it look less real.

In Lightroom or Photoshop I level the highlights and shadows a little. Then boost the contrast, saturation and vibrance. Toys and models tend to be painted in vibrant primary colours so I’m looking for the reds and yellows in the frame to really pop.

Because this was shot in RAW I have the option to alter the colour temperature and I warm it up in Lightroom to give it a lamp lit look. This will take some practice and each image is different. At this stage the image looks pretty bad as a standard landscape shot but hopefully it looks something like this.

With contrast, exposure and saturation adjustments
With contrast, exposure and saturation adjustments

Now into CS6 and onto the blur tools panel.  The first thing to do is copy the image to a new layer. This means that whatever adjustments you make, you have the original intact. Yes you can use the history panel but trust me, press cmd-j to create a new layer copy and work on that.

First we’ll try ‘Filter>Blur>Tilt-shift‘ you should be presented with a screen with a letterbox area over your image and a control dial in the middle.

  • The area you are keeping in focus can be moved by clicking and dragging the centre of the circle.
  • The size and rotation of the area can be adjusted by dragging the white dots on the middle lines.
  • The transition area from focus to blur can be controlled by dragging the dotted lines.
  • The amount of blur can be controlled either with the slider on the right or by clicking an area within the outer circle.
Using the tilt-shift filter
Using the tilt-shift filter

Make sure you have the ‘preview’ box checked top-centre to see your changes in real time. You can see that this tilt-shift effect works well for most of the image but we have a few problems. The biggest of these is the building on the right. Despite the whole building being the same distance from the camera we have a clear section half way up with the base and top being blurred. If we had no structures such as this in the frame the tilt-shift filter would work well. For this example it might be best to apply an ‘iris -blur’ instead.

‘Filter>Blur>Iris blur’ gives us a slightly different interface. This time the centre controls are the same but the area we are controlling now starts as an oval and can be stretched to fit a specific area. This tool does take some getting used to, by dragging the white dots around you can only change the size of the inner area. The shape must be controlled using the outer edge so if you need to stretch the oval wide make sure to press the ‘f’ key as soon as you open Photoshop to allow full screen viewing.  The iris is ideal for keeping the focus in the area we need with a setup something like this.

Using CS6's Iris blur to focus on one part of the image
Using CS6’s Iris blur to focus on one part of the image

Once you are happy with the result click ‘ok’ to render the final image. Don’t be afraid to push the contrast and colour and play about with different settings. Below are examples all shot in the same day. Some work better than others, for example the less details you can see on a person the better it can be. Some images have multiple zones of focus and some are very quick and simple.

I hope this tutorial has been helpful. Now go out and shoot!

Tutorial – Using layer masks in photoshop for a creative self portrait

I love my new studio, it’s a great space and a huge room that dwarfs the 3m backdrop. An empty photograph doesn’t do it justice, so how do I shoot it to give a sense of the space available? This is what I came up with, its a cool shot that’s not too difficult to achieve with the right setup and a methodical approach to Photoshop.

The Venue

In choosing your setting pick somewhere that will give you plenty of opportunities for different poses. The studio worked great because there’s plenty to do and things to adjust and look at. The more variation you can use the better the effect I think.

You can use as few or as many layers as you like.
You can use as few or as many layers as you like.

The setup

Your first step is to frame your picture and secure the camera to a sturdy tripod so It won’t move. At all. Most good tripods have a weight hook on the bottom. Hang something heavy (check the capacity of your tripod) on this to increase stability if you’re shooting outside. You must not zoom at all during your shots, you will have no problem with a prime lens or twist zoom. If you use a push-pull zoom like I did for this shot you might want to run a little tape over the barrel to make sure everything stays put. Once everything is framed you can consider your focus. Pick a spot to focus on at the distance you will be standing. Once you have your focus, switch it to manual and leave it alone. For the shots to match up the focus must be identical for every frame.

For my shot I knew I wanted everything sharp and printable on a large scale so at f16 and ISO 200, this gave me 1/10 second for the shutter. That’s fine at this focal length on a tripod as long as I don’t move. The tripod negates ay camera blur so my only limit is how long I can stay still.

Now make sure all auto settings are turned off. Auto ISO and Auto White Balance will give you variation across frames. Pick a white balance preset or value that looks good on your screen, if you shoot in RAW you can always alter this later. Each exposure must be the same for this to work so if you’re not sure try taking a photo with your hand close to and covering the centre of the lens. If the exposure looks the same and your lens doesn’t move then your camera is not adjusting for focus or light. Perfect.

Shooting

To make this easy on myself I used a radio remote to trigger the camera with a 5 second timer on the shutter. This gives me time to pose, press and hide the remote before each shot is taken. If you don’t have a remote trigger, just set the timer for as long as you need and be prepared to do a lot of running. Think carefully about where you can pose when framing your shot, if this is your first time using masks in Photoshop try not to overlap your images. Leave a decent gap in your frame between each pose to make the editing easier. Of course if you can do it, overlapping the images only adds to the effect and makes it more convincing to the viewer.

Take a shot of the empty setting and make sure your happy with the preview. Now the fun bit, snap away. If you plan your poses carefully you can interact with yourself, experiment to see what works for you but be aware of reflections and shadows. In my shot the reflections in the floor and shadows on the seats all add to the realism. If you aren’t too comfortable with Photoshop make sure these don’t interfere with each other or overlap.

When you think you have enough options go through your shots on camera but try not to move anything. If you want to retake any pictures you’ll never get it lined up exactly right. If you see any you don’t like go back and take some more. Better to have too many options than too few.

Image processing

If you’ve done everything right, you now have a set of images identical except for your position in them. If you want to make any global adjustments such as saturation, contrast or levels you have two options.

  • Compose your photos into one image before adjustment then treat the resulting image as a single photograph to edit, possibly better if you shooting JPEG but not for me.
  • Make global adjustments in software such as Lightroom or Camera Raw before importing them to Photoshop.
Right click and select 'open in layers'
Right click and select ‘open in layers’ in Lightroom

I always choose option two as Adobe’s Lightroom software is perfect for group image changes, once imported I can work on just one image until I’m happy with the result, then synchronize my settings across all the rest. You can do this with Adobe’s camera Raw too, a plug in used automatically for RAW files. If you are shooting JPEG right click on the your selected files and click ‘open with’ then select adobe camera raw to access this easy adjustment and sync for your images. These steps are for Mac OS users but I’m sure windows has a similar command.

Using 'Place' in Photoshop
Using ‘Place’ in Photoshop

So, assuming your familiar with basic editing, first adjustments are done and your ready to start composing. You can skip straight to this step if your happy with your exposure. If the image is good straight from camera go ahead.

In Lightroom select the images you want to use, let’s say you use two, one picture sitting on a chair, one standing up next to it. Right click (ctrl click) to bring up the menu. Select ‘edit in’ and then ‘open as Layers in Photoshop’ to let the two programs import them all into your layers window for you. If you’re using Camera Raw or just Photoshop, open your first image then in the File menu select ‘place’. This will create a new layer containing your second image, directly over your first. for large numbers of images repeat until you have them all imported, each in their own layer.

Make sure to name your layers
Make sure to name your layers

At this point it’s a good idea to label each layer. Make sure you have your layers window open (F7) and double  click on the layer names to edit. If you have trouble identifying which is which, click the eye icon on the left of any layer to toggle it’s visibility or alt-click to see only that layer. When each layer is named, for example ‘sitting/ standing’ alt-click the eye icon next to the first layer (sitting) so only that layer is visible. This is our starting point for the first layer mask.

Creating a 'Reveal all' layer mask on Layer 2
Creating a ‘Reveal all’ layer mask on Layer 2

Using Layer Masks

Click the eye next to your second layer (standing). We want to bring in only the parts of this image that are different from the first so we use a layer mask to choose the area to keep. Click the second layer so it is selected and in the Layer menu highlight ‘Layer Mask’ and click ‘Reveal all’

Mask as shown in the Layers window
Mask as shown in the Layers window

Nothing happened. The ‘Reveal all’ has applied a mask that is currently blocking the pixels underneath at 100% opacity. You can see in the layers window that ‘standing’ now has a new white box next to it indicating it is linked to a layer mask. As long as that layer mask box is selected we can tell the mask how much of the image to let through by painting onto the mask in shades of white, grey and black.

Select the brush tool in your palette and select black as your colour. Use a large soft brush to fade the effect and avoid harsh

Use a soft black brush to erase yourself until there's no visible transition
Use a soft black brush to erase yourself until there’s no visible transition

edges if there are any differences between your layers. To change your brush size and hardness just right click anywhere on the image. Paint over the area in your frame you wish to keep, until you have erased yourself completely from the second layer, as you do so you will see a mini version of your work in the layers window. Your image should now look exactly like the first layer with no harsh edges from your brush or evidence that you were ever there. And then.. make sure the layer mask is still selected and press cmd-I to invert the mask.

Now there are two versions of you in the image with a seamless transition between them. Note the box in the layers window has flipped from black on white, to white on black when we inverted it .To bring more into the picture from layer 2 use a white brush, to trim it down use a black brush. You could skip straight to here by selecting ‘Hide all’ in the Layer menu earlier instead of ‘Reveal all’ but this does mean guessing your shape and positioning from a blank canvas when painting with the white brush. I find it easier this away round.

With the layer mask selected, 'invert' to see the result
With the layer mask selected, ‘invert’ to see the result

For multiple images that do not overlap, you can keep repeating this process. When you are happy with two layers. Click the eye icon next to the third. Select that layer and create another layer mask. If you’re images do overlap you will need more than the brush to get a convincing result but that would make this post even longer and you’ve already done very well to get this far so try this first and we’ll talk about quick masks another time.

Once you are happy with your image you’ll need to save and export it to JPEG format for normal use. Save your PSD file now while the layers are all clear and easy to edit. This may take some time if you have 11 large files in masked layers like I did but do it anyway. With a PSD file you can re edit each layer as much as you like. Only delete this when you have your finished product in saved in the highest quality you need.  If you are completely happy and ready to ‘save as’ a JPEG, then right click on any layer in the layers window and select ‘flatten image’ this reduces the file size and makes it a much easier job for Photoshop to export and save.. If you opened your files through Lightroom. Closing Photoshop and selecting ‘save’ will create a TIFF or PSD file in your source directory and add it to your library automatically. You can then export it straight from Lghtroom as normal. Clever stuff.

Photoshop tutorial – How to turn your model into a marionette puppet

A little while ago, I posted an image of Nat, a previous model of mine I turned into a puppet. I’ve had a suprising number of messages asking how it is done and though I have no gone into every detail, I have outlined the steps that should get you started turning your own friends into puppets.

To start with you’re going to need the right photo. Obviously the positioning contributes a lot to the final image so I asked Nat to give me a broken doll pose. Shooting against a plain background of black or white will make the first try easier, once you have mastered cutting images out you can start transferring your puppets into other backgrounds but as this will involve some complex cloning it’s best to start with a block colour background. If you want to shoot outside or against a specific backdrop use a tripod for your shot and take another image of your set without the model. It it much easier to bring in the background from a reference shot than it is to create it!

If you can’t wait to get started and would like to use this image and get started straight away you can download this jpeg here.    This is an old snap from a 6mpx D70 but it serves well for the purposes of this tutorial.

Remember as you read that there are many ways to achieve the same results with photoshop. Here is just one method to get results quickly and easily. Aimed at those who are fairly new to photoshop techniques. We will use the smudge, shape, dodge, burn and brush tools for this effect. All steps can be completed with alternate methods so use the tools that your are familiar with.

Step 1 – Skin smoothing

For this image we want to make the skin look smooth and plastic like. When we smooth skin for portraits we use masking and layer effects to preserve skin tone. Here we want an artificial look so try using the smudge brush with hardness set fairly high. Use small brush strokes and paint along the limbs and skin and watch the texture turn smooth without losing the colour range. This method is much quicker than masking. If you find the effect to strong, use a low opacity on your brush and go over the area multiple times until you get the right look.

Plastic skin with the smudge tool

Step 2 – Dodge and burn skin

This lighting for this image is quite flat so you may not need this step. Here it really adds to the rounded plastic feel of the limbs. Start with a the dodge tool, zero hardness with an opacity of 5-15%. Set your brush to dodge only highlights and go over the middle of the limbs to provide the shiny highlights. Use a smaller brush to go over the cheekbones, nose and chin. Once you are happy with the result we’ll add some shadows. Next switch to the burn tool, again with a low opacity but this time set to burn the shadows. Use this brush to add shadow to the outside of the limbs, don’t be afaid to be bold with your contrast, this only adds to the toy effect. Frequently zoom out as you work to check how the image looks as a whole.

Dodge and burn to create depth

Step 3 – Removing the joints

This is no doubt the trickiest part. We will be removing a section of the limb and painting in a background. First use the eyedropper to select a colour close to the skin tone, then with the ellipse tool draw an ellipse to represent the flat end of  limb. Then use the transform command (apple/cmd T) to rotate and stretch your end shape into position. Now we must either paint in a black background with a brush or if you are familiar with masking, copy your background layer (apple/cmd J) then paint over your background in black and mask off the remaining area until you have the correct gaps in the limbs. There are as many tutorials are there are different ways to manage masks, if you have trouble with this step why not book a one to one photoshop session. When using a detailed background you will need to create a false image and mask through your original. Start with block colours.

Paint or mask the background to create gaps

Step 4 – Connecting the limbs

You can be as creative as you like here, use robotics, wires, bearings and joints. This is the simplest way to connect the limbs as if its just a single piece of bendable wire. Start with a medium grey circular brush at 100% hardness and opacity. On a new layer draw a small connecting line from the middle of your ellipse shapes and over the gap we created in the last step.  Then simply use the eraser tool, or create a masking layer to clean up the overspill. Don’t worry that it looks flat at this stage, once your wires are drawn and cleaned up, use the dodge and burn tool exactly as we did to the limbs in step 2. Zoom in and use a small brush with low opacity and hardness to first add highlights to the centre of your wire and shadows to the edges. Repeat this step until the wires appear to have depth even when viewing the whole image.

Painting in the connecting wires

Step 5 – Adding strings

Again there are many ways to do this, for the full look import images of strings or chains and mask away what you don’t need. For this simple image select a light grey with the line or pen tool and start to draw your strings from as many anchor points as you wish. Remember to select ‘stroke path’ with the chosen tool otherwise you’ll just be selecting a path. Once your lines are drawn create a layer mask to delete the strings where they pass behind the body. The more complex the body shape the more work you’ll have to do here. If you use a thin string you can get away with them attaching directly to the skin. If you use a thick string try adding a ‘bevel and emboss’ layer style to quickly achieve a realistic look.

Adding the strings

Step 6 – Final touches

The last thing to make an instant marionette is that creppy dummy mouth. Use the pen tool or polygonal lasso tool to draw around the top of the lower lip and straight down to the chin. Once this selection has been made, copy the contents from the background layer with (apple/cmd C). Now use the eyedropper or palette to select a dark reddish brown and with a soft brush paint into your selection. Because you have an area selected you can draw a clean shape without effecting the detail we want to keep outside the selection. Once you are happy with your new mouth, deselect the area (apple/cmd D) and paste the chin back on top (apple/cmd V). Drag the mouth down a little to reveal your handy work.

At this stage go back to your dodge and burn tools. Dodge and burn are your friends. Always making little changes with a low opacity add depth to your joints, wires and anything you have painted/created until you are happy that it does not detract from the main image.

All finished with final dodge and burn

That’s just about it. Now look away from the screen, make a cup of tea, have some toast or something. Come back and look at the image again in a few minutes. You’ve been working on it for a while so try to see it afresh. Now is the time to clean up any details you’ve missed before selecting all layers and hitting merge layers (apple/cmd E) to create a single layer before saving. If you want to come back to work on the separate layers be sure to save as a .psd first.

This image was produced in photoshop CS6 but should apply to all versions.

Let me know how you get on, or what you’d like to see next by leaving a comment. Why not post a link to your version and show everyone how you did?

Photo retouching and skin smoothing. How much is too much?

Quality skin smoothing should enhance, not replace makeup

I am often asked by clients, both portrait and wedding alike – “can you airbrush me?” The answer of course is “yes, but do you really want me to?”

When it comes to the controversial topic of photo editing – specifically skin smoothing known as ‘airbrushing’ it can be difficult to tell what is acceptable in the fast moving media industry.

Do a quick search for “airbushing” and you’ll be met with a volley of opinions both in support of, and rallying against this well established technique. First it is important to note the difference between the different levels of photo enhancements ranging from simple blemish removal to full shape-changing manipulation.
For top models and celebrities, airbrushing is just another part of the daily routine with some even employing full time photo editors to ensure continuity from one shoot to the next. For clients looking for a casual photo shoot it can be used to enhance the natural look and provide that little extra to provide images you’ll want to treasure forever. Many models walking into the studio are basing their expectations on magazine work, with hours of professional hair and makeup, some minor tweaks in the editing can make a huge difference to the end result.

Here’s how I catagorise the different levels of editing.

  1. Blemish removal – This is practiced by virtually every photographer. The aim is to remove only temporary blemishes such as spots or scratches. Easily achievable with a variety of software this does not change the overall look or texture of the skin.
  2. Asymmetry removal – This is a dangerous area, referring to any marks or blemishes that are a permanent feature. Photographers must be very careful when removing moles or scars. It’s easy to cause offense by removing something you think the client wants rid of before asking.

    Here the lines have been smoothed only slightly to remove the pinching caused by tight fitting lingerie.
  3. Light skin smoothing – This method uses a simple blur filter and some photoshop masks to produce a texture free skin. This method tends not to stand up to close inspection and is easy to spot. ‘Smooth’ and ‘blur’ are not the same but it can be achieved quickly to give a client a good overall impression of airbrushing results.
  4. Complex skin smoothing – Using a variety of techniques such as channel mixer, high pass, layer styles, masks and much more (I don’t want to give up the secret recipe in it’s entirety) skin can be made smoother but retain the texture giving a look of high quality make up rather than photo manipulation. This is my preferred choice as it is more flattering and less noticeable. If I am smoothing skin for a beauty shoot, I don’t want people thinking ‘look at that editing’. I want them to think ‘look at that model’
  5. Post process lighting – This is an interesting one. By adding highlights and shadows to your images you can change the way a model looks. By lightening the area under the eyes and darkening the cheeks you give the appearance of high cheekbones for example. Again this is a dangerous game as you risk changing the look of your model too much. A little can go a long way here…
  6. Manipulating shape – This is the part that really fuels the media frenzy. Changing the shape of a model to ‘enhance’ features by shrinking some parts while enlarging others, giving celebrities the unachievable figure made so fashionable by the glamour and music industry. This is done (often very obviously) by using tools such as ‘liquify’ or ‘puppet warp’ and should be approached only with the full consent of your model. Changing someone’s shape because you think it could be better is not a always a good enough reason. That said there are times when minor adjustments can make a huge difference – smoothing out the lines where clothes pinch against skin can reduce distractions in your picture or remove creases in material that would otherwise pull attention away from your model.

And now I ought to come to some concluding point that says ‘do it this way and you’re safe’. Unfortunately no such method exists. The key is to be sensitive to the needs of your client. I make sure that models and clients know exactly what they want from their finished shots and where they are heading. For a beauty shoot destined for facebook, the aim may be to make the client look the way people think models ‘should’ look. Typically this means a little smoother, towards the controversial end of the scale. For acting head shots however it is important to show the client in their best light, but without removing any features that would impact upon a role once cast.

If you are starting out down this road maybe these pointers will help.

  • Save your work continuously, keep a copy of the original shot to hand for reference. If your model looks like a different person, you’ve gone several steps too far.
  • Know how your images will be viewed. A full body shot on facebook will never be viewed at 100%, so editing at this level will go largely unseen.
  • Conversely, if your images are for large format print, every detail will come out so watch out for transitions between smoothed skin and fine details.
  • Be honest and open with your model. Explain the options and provide examples if available. A simple question like, ‘would you like me to do any enhancements’ is much better than ‘do you want me to fix this?’

If you would like to learn more about post processing using Adobe’s creative suite, get in touch for photographic tuition from Jon Hall. Sessions can be conducted in the studio with everything to hand from camera to print or in the home with your setup to make the best of what is available to you.
If you would like to have your say, leave a comment on this post. How much is too much?

Turn your model into a marionette

Ever felt like puppetifying your models?
Photoshop offers way more than corrective adjustments. Using a few simple techniques you can turn your favourite models into puppets and create this cool (if a little creepy) effect.
If you’d like to see a tutorial to guide you through the process of creating an image like this leave a comment and let me know.

Turn your model into a puppet
Learn the puppet effect on photoshop, just leave a comment.

Photo manipulation – toy town project

Known as tilt shift, this effect traditionally relies on specialised lenses to produce the toy town effect you see here. Now these amazing results are available to us through the use of editing tools such as Adobe’s photoshop software. Click here for details of photo editing and manipulation courses.

This slideshow requires JavaScript.