Skip to main content

I smell photoshop

While recently browsing cheap tripods on amazon (you can never have too many tripod when doing camera trap type work...) I stumbled on one which included this image:


Instantly I was drawn to something that made me suspect this advertisement had been photoshopped! How dishonest! Now, you might spot the giant hand next to the man and assume that is what I am referring to. Not so. Zoom in a bit closer to just the model. 


The tripod on the backpack just looks wrong. Note the lighting on the left tripod leg in particular, very different to the fairly diffused light on the model. The colour is also wrong compared to the rest of the image. Pick out areas that you are fairly certain will be black, for example the lens hood. Does it match the black of the tripod? No. It is almost a greeny blue while the tripod appears very black. The tripod is much warmer than the rest of the picture.

Can we prove the tripod has been added digitally?

There are various techniques to try, such as Error Level Analysis, but in this case the results are pretty subtle, or at least they are to me:


Another option is to try and find the original image, assuming they have just used a stock photo to try and sell the tripod. You could google for terms like "man with backpack taking photo in snow" but you'll quickly find that route isn't sensible. Another option is to use a Reverse Image Search like TinEye.

Since we are making the assumption that the tripod has been digitally added, we need to search for the image without the tripod and the easy way to do that is to crop just the model and upload that to TinEye.

Oh. Epic failure. So what now?

One observation is that the model appears to be using a left handed Nikon (you can see his left finger appears to be on the shutter button) so therefore the image has been flipped. Lets try mirroring it.


That's more like it. We can see this is indeed a stock image and it doesn't contain the tripod. Hurrah. We've successfully proven something that everyone could tell but nobody cared about anyway, that an advert for an inexpensive tripod on Amazon photshopped a stock image rather than paying a model to demonstrate it in a beautiful snowy landscape. 

I'll admit I was impressed when TinEye managed to find a cropped and heavily reduced version of the image, so if this post has anything to offer it is that little observation. Maybe it will be useful for your own analysis one day.

Labels:

Comments

Post a Comment

Popular posts from this blog

Experimental macro flash diffuser

The abundance of insects over the last couple of months has been a perfect opportunity to experiment with different macro flash setups. Using a standard speedlight as-is isn't really an option, as the light is projected out much further than the front of the lens (which is exactly what you'd normally want a speedlight to do) and is much too harsh. There are therefore two problems that need to be solved; diffusing the light and directing it closer to the front of the lens. There are of course multiple commercial solutions available, such as the Canon MT24-EX or the Sigma EM-140 ring flash, but these are all relatively expensive and unlikely to be an option to anyone wanting to experiment with macro photography. A speedlight however is much less of a barrier to entry. A few different prototypes were made but the design I settled on was a light pipe style design that captures the flash light and funnels it down to the front of the flash. Of course, it would be ideal to hav
Post a Comment
Read more

Macro diffuser update

I've simplified the macro diffuser that I open sourced last year . The MK1 version required printing out a custom lens hood which would only attach to a Canon 100mm macro lens. The MK2 version no longer requires a custom lens hood, it simply clicks on to the hood of whatever macro lens you might have making it far more versatile and easier to print. The tested lenses so far include: Canon 100mm  Canon 180mm Sigma 180mm Sigma 105mm This hood helps reduce harsh shadows you often get when using flash but the main advantage for me is that the head can be positioned depending on how close you get to your subject. This is the typical kind of result I'm getting with this setup now. To build your own version you just need to print the model and supply a few basic parts: M3 x 40mm Hex Head bolt (the shape of the head is hex, NOT the shape of the tool used to tighten it). M3 wingnut. Diffuser material. I use the plastic casing found on some notebooks and it seems t
Post a Comment
Read more

The laser trigger module overview

The laser trigger module is the most useful gadget to come out of the camera trigger project so far. It is a simple board that controls up to two lasers with sensors to detect if the beams have been broken. The electronics involved in detecting a broken beam are very simple so using a microcontroller might at first glance seem to be unnecessary overkill, but using a microcontroller gives a significant amount of flexibility in a small and inexpensive package.
Post a Comment
Read more