Even if the Camera Never Lies, the Retouched Photo Might…

In Hyper-reality Offline – Creating Videos from Photos, we saw how a single flat image could be transformed in order to provide a range of video effects. In this post, we’ll review some of the other ways in which photographs of real objects may be transformed to slightly less real – or should that be hyper-real – objects, and consider some of the questions such manipulations raise about the authenticity of photographic images.

In 2014, an unsuccessful bill was introduced to the US House of Representatives that sought to introduce controls around “photoshopping”, based on the principle that “altered images [of models’ faces and bodies] can create distorted and unrealistic expectations and understandings of appropriate and healthy weight and body image” (Slate: Legislating Realism). The Bill reappeared in the 114th session of Congress in 2016 as H.R.4445 – Truth in Advertising Act of 2016:

This bill directs the Federal Trade Commission (FTC) to submit a report to Congress assessing the prevalence, in advertisements and other media for the promotion of commercial products and services in the United States, of images that have been altered to materially change the appearance and physical characteristics of the faces and bodies of the individuals depicted. The report must contain: (1) an evaluation of the degree to which such use of altered images may constitute an unfair or deceptive act or practice, (2) guidelines for advertisers regarding how the FTC determines whether such use constitutes an unfair or deceptive act or practice, and (3) recommendations reflecting a consensus of stakeholders and experts to reduce consumer harm arising from such use.

OPTIONAL READING: Forsey, Logan A., “Towards a Workable Rubric for Assessing Photoshop Liability” (2013). Law School Student Scholarship. Paper 222. http://scholarship.shu.edu/student_scholarship/222

Photoshopping – the use of digital photograph editors such as Adobe Photoshop – is a form of reality mediation in which a photograph is transformed to “improve it”. In many cases, photoshopped images may be viewed as examples of “hyper-reality” in that no additional information over and above the scene captured in the photograph is introduced, but elements of the scene may be highlighted, or “tidied up”.

As you might expect, the intention behind many advertising campaigns is to present a product in the best possible light whilst at the same time not misrepresenting it, which makes the prospect of using photo-manipulation attractive. The following  promotional video from the marketers at McDonald’s Canada – Behind the scenes at a McDonald’s photo shoot – shows how set dressing and post-production photo manipulation are used to present the product in the best possible light, whilst still maintaining some claims about the “truthfulness” of the final image.

Undoubtedly, many food photographers manipulate reality whilst at the still time being able to argue that the final photograph is a “fair” representation of it. In a similar way, fashion photography relies on the manipulation of “real” models prior to the digital manipulation of their captured likenesses. A behind the scenes video – Dove Evolution – from beauty product manufacturer, Dove, shows just how much transformation of the human model is applied prior to a photo-shoot, as well as the how much digital manipulation is applied after it.

Let’s see in a little more detail how the images can be transformed. One common way of retouching photos is to map a photographed object, which may be a person, onto a two dimensional mesh. Nodes in the mesh map on to points of interest in the image. As the mesh is transformed by dragging around nodes in the mesh, so too is the image, with points of interest tracking the repositioned nodes with the image content in each cell, or grid element, of the mesh being transformed appropriately.

As well as transforming body shapes, faces can be retouched in a similar way.

Surrounded as we are each day by commercially produced images, it’s important to consider the range of ways in which reality might be manipulated before it is presented to us.

EXERCISE: even if we may be a little sceptical around claims of truth in advertising, we typically expect factual documentaries to present a true view of the world, for some definition of truth. In 2015, the BBC were called to account around a documentary that appeared to depict a particular sort of volcano eruption, but that was actually a composited sequence from more than one eruption. See if you can track down one or two reports of the episode. WHat charges did the critics lay, and how did the documentary makers respond? What sort of editorial guidelines does the BBC follow in the production of natural history film-making?

EXAMPLE RESOURCES:
http://www.independent.co.uk/arts-entertainment/tv/features/bbc-feels-the-commercial-chill-of-fake-documentary-6276155.html
http://www.bbc.co.uk/editorialguidelines/guidance/natural-world/guidance-full
http://www.theguardian.com/media/2015/oct/03/bbc-lightning-erupting-volcano-fake-patagonia
http://www.bbc.co.uk/earth/story/20151002-how-to-capture-a-volcano

EXERCISE: As well as advertising and documentary making, news journalists may also be tempted to photoshop images for narrative or impact effect. Watch the following video, stopping and pausing the video where appropriate to note the different ways in which the photographs reviewed have been manipulated. To what extent, if any, do you think manipulations of that sort would be justifiable in a news reporting context? 

A post on the Hacker Factor blog – Body By Victoria – describes how a photographic image may be looked at forensically in order to find evidence of photoshopping.

As the Photoshop tools demonstrate, by mapping a photograph onto a mesh, transforming the mesh and then stretching pixel values or in-painting in a context sensitive way, photographic images can be reshaped whilst still maintaining some sort of integrity in the background, at least to the casual observer.

Increasingly, there are similarities between the tools used to create digital objects from scratch, and the tools used to manipulate real world objects captured into digital form. The computer uses a similar form of representation in each case – a mesh – and supports similar sorts of manipulation on each sort of object. As you will see in several other posts, the ability to create photorealistic objects as digital objects from scratch on the one hand, and the ability to capture the form, likeness and behaviour of a physical object into a digital form means that we are truly starting to blur the edges of reality around any image viewed through a screen.