I chose this Image because I believe it would be a rather simple, yet effective composition that would not only display my various skills that I have learnt on the course within Photoshop, but also would include a nice array of colour, and lighting.
I began the process by using a Canon EOS 1100D, to take my very first images. I decided to do various angles, at different heights and lengths to ensure that I had the potential to get the right image to composite. I started taking photographs in manual, but unfortunately had issues with the focus, and wasn’t able to work out what I was doing wrong, therefore I had to use Auto Mode on the camera. The images I took were shot in RAW Mode, with a Bit Depth of 16 & Colour Mode RGB. To take these images, I had to measure the baseline measurements, along with the height of the camera as well as the angles of the photo’s were taken. This was a mundane task, but vital in terms of creating an accurate composition.
Here are examples of the different photographs that I had taken using the Canon EOS 1100D…
From these images, I decided to narrow it down to three of my favourites, then analyse them to see which would be the most suitable for my composition.
These were the three favourite images that I had taken…
This was the image that I chose. I believe that it had the potential to challenge me on my lighting skills in the green screen room, as well as getting my angles and measurements correctly. Doing these two correctly, would allow Photoshop to become easier for me, as I am not very experienced on the program, and would have difficulty in getting the image correct if I hadn’t taken the photograph to a great standard.
Here below is the META DATA & EXIF DATA for the original background image that I chose…
Here are the measurements for this image that I had made whilst setting up to take the photograph :
 |
| Image Not To Scale |
Once having the ideal background image for my composition, it was time for me to head to the green screen room to take the images that I was going to scale down. As I am not very keen to be in front of the camera, I managed to get a good friend of mine Giorgia Perini to be my model for this task, she was great to work with, f which I very much appreciated!
All images that I took were in RAW MODE, by also using manual focus, ISO & manual Exposure.
As I was not using any flash on the camera, It was important for me to get the details right to have a decently exposed image.
To do this, I had to ensure that I had positioned the lighting that I had to work with, in the correct area’s of the green screen room, to replicate any shadows that were to be seen in my original background photo. I did have difficulty in using my previous measurements, but I feel that overall it worked out well in the end.
The idea was to make the image look as if Giorgia was hiding behind one of the cards, therefore parts of her had to be hidden from view of the camera. To do this we had to improvise a little, and I decided to use one of the green screen floorboards representing one of the playing cards. Although entirely improvised, it worked well, and was a great prop to work with.
After taking multiple shots on the camera, It was time for me to reflect on the images that I had, to see if they were suitable. These were the shots that I had taken during the shoot in the green screen room.
I eventually came to the decision that I would use this following image as my one to composite with the backround image that I had taken. Below is also the METADATA & EXIF DATA for this image.
After completion of the greenscreen photos, it was time for me to display my skills in Photoshop.
Therefore I opened the RAW image that I was happy with into photoshop ready for editing.
The first step for me to achieve the image I wanted to create was to KEY the image. I did this by using the channels palette, selecting the appropriate channel from the RGB, and going to the Image drop down menu, and selecting the calculations button.
From here I had to select the appropriate layer from the dropdown menu, as well as the appropriate channel that I was either going to SUBTRACT, ADD or MULTIPLY from. This task had to be repeated several times, using different blends, Such as Subtract or Multiply to get the desired colours chosen.
At this point, you can see that I had Green Spill within my image that I had to get rid of.
I would get rid of it by using the technique that we learnt in our course called de-spilling. I started off by creating a copy of the layer chosen, then removing the layer mask from it. Then i transferred the layer mask from the original layer, by sliding it up to the new copy. I then renamed the layer as De-spilled Layer.
Going back to the channels palette, I then used the curves button and had to half the output on the Red and Blue channel by 50%, which is the figure 128. To do this I made a copy layer of each the blue channel and the red channel, then used the curves button to lower the output.
From this, I then went onto combine the red and blue channels together by selecting one, and using the calculations option in the drop down menu. From there I used the Add blend mode which then combined both channel layers together, which gave me my average red and blue.
With my average red and blue completed, I followed on to the next step which was to subtract this new Average RED and BLUE from the original green screen layer. This was done using the same calculations method, only using the subtract blend mode instead. From that I had created my spill map.
From this point I subtracted all the green, by using the same technique as before, by using the appropriate channels layer. I copied the new de spilled green layer, and pasted it onto the green channel. This worked pretty effectively in terms of eliminating the green from my image, but still I had a few issues with it that I had to sort.
I sorted the issues out by using the eraser tool, which usually isn't the correct method, but it worked for this particular image.
Once I had finished with the despilled layer, it was then time for me to combine it with my original background image of the playing cards that I had taken. This took some time as I did have to change the scale, and also rotate it a little for it to fit in with image correctly.
When I first applied the De-spilled layer, the very first thing I noticed was that the colours didn't match correctly with backround. Therefore I had to change the colours to make it fit seamlessly. I did this by using the Vibrance tool, which allowed me to change the saturation of the de-spilled layer.
BEFORE COLOUR CHANGE
AFTER COLOUR CHANGE
After getting the positioning, the scale and the colour to match the backround, the next step for me to do was to create a shadow that would be realistic for the little person. I had to acknowledge the fact that in the backround image the shadows were all facing the left hand of the picture, therefore I had to match my new shadow to the same direction as they were following.
To do this I copied the original dispilled layer, then used various adjustment layers and the saturation tool to get rid of any colour that existed. I then had to flip the image horizontally then vertically therefore it would make a fairly accurate shadow. I then had to rotate it to match up with the other directional shadows, so it looked genuine.
By this point I was relatively happy with the outcome of what I had, but decided to ask advice and opinion of our lecturer for this module. From this, he brought up some very valid points on how to improve the image.
From the feedback, he mentioned about the colour temperature of the image, as the girl wasn't matching up to the rest of the photograph, therefore it needed to be toned down to a cooler colour.
He also mentioned about a reflection that I had missed out, which involved the foot of the girl. As the surface she is standing on is reflective, in a realistic situation, her foot would be slightly reflected directly where the shadow appear's , therefore that was something I should rectify.
Another point that was brought up, is that when looking close up onto the girl the image doesn't quite blend fully with the background. The original background had far more natural grain, where as the super-imposed image did not. Therefore I made sure to add more grain to the image of the girl so it would blend seamlessly into the image.
The final point that was brought up, was that he would like to see what the girl was afraid of, and what she was looking at. Therefore I had to take one last final image to composite into it. I decided I was to incorporate a large face looking down on her, to emphasise on how small the girl is.
Here is the EXIF data of the new image that I was to incorporate into the composition...
For this image, as I was short on time I wasn't able to use the green screen to take the image, therefore when it came to editing I had to use the quick selection tool to remove the background. I noticed instantly that the colours weren't accurate enough to fit seamlessly in the composition. I had to use the channel mixer along with the vibrance tool to make it more accurate. I also skewed the angle of the image to make it seem like she is looking down on the girl. To get it as close as the original image, I used a final levels tool just to create a slight shadow underneath the chin area, with more light on the nose.
By doing so I managed to get this outcome below...
As I was rather pleased with the outcome, I decided that I felt that I only needed to add final touches to the image, the minute details.
Some of these details included creating a reflection onto the pack of cards that was in the foreground as it was naturally reflective. So as you can see in the image below, I added a subtle layer over the image that was flipped horizontally, and vertically, then with the opacity and transparency on a low figure. This can be seen in the image below...
This can be seen in the image below...
No comments:
Post a Comment