Jun 6, 2015 - hBp://www.dailymail.co.uk/sciencetech/ar]cle-â2761272/This-âNOT-âreal-âwoman-â. Meet-âBeryl-â
This is a composite of a photogrammetric rendering of this ta2oo of my grandmother’s carpet My father dedicated his phd in computer science in the 60s to his mother As teaching him how to code by watching her knit
1
Presenta?on at 21st century photography: art, philosophy, technique 5-‐6 June 2015 University of the Arts London h2ps://photoconference2015.wordpress.com/
2
I want to describe the emergence of forms of surface between digital photographs that has been mediated by computer vision I want to set out an argument that these surfaces have the poten?al to reshape the structure of thought. And I want to propose that MESH is both a technological artefact of the structure between photographs and A resonant conceptual model
3
my PhD crea;ve prac;ce has been framed by a ques;on on the nature and cons;tu;on of surface in digital photographs. Specifically: Given the dematerialisa/on of the photographic image, to what extent can a photograph be regarded as having a surface? The short answer is yes, there are many forms in which surface is present in digital photographs. • Photographs have a fundamental rela?onship with surface. • A photograph may be conceived as an impression of surface -‐ of light reflected off surface; • a photograph is itself a surface – a two dimensional plane; Be it a physical print or a mathema?cal concept • and the screen may be conceived as a form of surface. I have also come to the posi?on that ‘dematerialisa?on’ is NOT a given
4
Surface is simultaneously a material, abstract and psychological en;ty. Surface is the place of contact and separa;on, transfer and shedding, the boundary of expansion and contrac;on. It is a powerful guide by which to interrogate an en;ty, an environment within which we are immersed. In this context, I propose to employ it as an ontological metaphor with which to visualize a shiD in shape and boundary. Surface and depth are co-‐crea;ve and indivisible And I am delighted that surface appears to be one of the emergent themes of this conference.
5
Whilst telling this story, I will be observing two intertwined contradictory trajectories: • That technologies create paradigms • And that technologies are created to meet desires
6
I am drawing on Flusser’s model of how technologies shape thought. In summary, he proposed that • the wri2en word facilitates linear thought; • that photographs are surfaces, two dimensional, and facilitate scanning modes of thought (Flusser, 2000); • whereas computer technologies create interconnected networks that facilitate more complex thinking I wish to consider this model in the context of An algorithmic environments for digital photographs.
7
As has Batchen observed whenever somebody invents a new form of imaging technology, the first thing they do is make an image of their child. Perhaps for precisely, imaging technologies are created BECAUSE we want to capture images of our children. The impulse is emo?onal and personal Fox Talbot’s disappoin?ng drawings of his honeymoon could also be classed with these. Morse’s inven?on of morse code and the telegraph system following the death of his wife One aspect of the contested place of photography can be characterised as a tussle between those who view the medium as inherently ?ed to certain technologies and those who define photography as a specific set of prac?ces, as photographic impulses or desire (Batchen 1997 p.212, Maynard 2010 p.29, Warner Marien 2012, p.6 ). In his
8
Whilst images have always been dialogic, in dialogue with and connected to other images, this has been amplified and made more tangible by the algorithmic turn. As a result, it is possible to visualise the rela;onships between images has having geometry, form and, therefore, surface. Specifically, I want to consider this in two forms The first is photogrammetry – reverse image search engines -‐ such as Google ‘Search by Image’. Both these applica?ons of computer vision draw out rela?onships between images that have both surface and depth.
9
Art+Com undertook some significant visualiza?on work in 1995 that gave shape to moving image sequences. Rather than posi?oning the screen as an invisible immobile portal, The Invisible Shape of Things Past described the movement of the camera image through space and ?me. The resul?ng forms were 3D printed. These works demonstrate a rela?onship between space, surface, ?me and movement Joachim Sauter et Dirk Lüsebrink, Invisible Shape of Things Past Serial vs paralell – Gallaway Internals between image planes can measure ?me but can also measure space h2ps://vimeo.com/95422036 34”
10
Art+Com undertook some significant visualiza?on work in 1995 that gave shape to moving image sequences. Rather than posi?oning the screen as an invisible immobile portal, The Invisible Shape of Things Past described the movement of the camera image through space and ?me. The resul?ng forms were 3D printed. These works demonstrate a rela?onship between space, surface, ?me and movement
11
The CCD sensor records a grid of measurements of light intensi?es Sample points It is this data that facilitates what is known as computer vision but a computer does not ‘see’ in the cogni?ve sense that a human subject perceives through vision. Rather, computer vision is the func;on of a set of algorithms that automate a compara;ve search for a match between a paNerns of data (Turek, 2011). Computer vision, also known as machine vision, has found applica?ons in a range of selngs, such as factory automa?on and automated naviga?on. I want to limit considera?on to some of the ways in which computer vision is impac?ng on photography.
12
s?tched panormas were my first encounter with algorithmic image manipula?on that employed a form of computer vision.
13
Working with overlapping digital photographs of interiors, the somware iden?fied the matching overlapping elements within the individual captures and ‘s?tched’ them together to create panoramas
14
Both these images are built from the same data, the same set of captures Whereas the s?tched panorama fla2ens and unwraps the scene, photogrammetry calculates the rela?ve shapes of the content -‐-‐-‐-‐-‐ Mee?ng Notes (6/06/2015 14:37) -‐-‐-‐-‐-‐ 2015
15
Not the full set
16
Images aligned and distorted via computer vision
17
Interested in the imperfec?ons, the wonky bits, the messy density of the meshes And the gaps lem where the program perceives blank spaces. Will be exhibited via a virtual reality headset You will be inside the virtual room whilst inside the gallery photogrammetry is defined by Kyle as: … methods of image measurement and interpreta/on in order to derive the shape and loca/on of an object from one or more photographs of that object. In principle, photogrammetric methods can be applied in any situa/on where the object to be measured can be photographically recorded. The primary purpose of a photogrammetric measurement is the three-‐dimensional reconstruc/on of an object in digital form […] or graphical form (images, drawings, maps). (Kyle et al., 2013) FYI, this object will be exhibited in the Tinning Street Gallery via a virtual reality headset, probably Google Cardboard
18
Imperfec?ons -‐ interested in the wonky bits, the messy density of the meshes Will be exhibited via a virtual reality headset You will be inside the virtual room whilst inside the gallery
19
Photogrammetry literally demonstrates the emergence of a surface formed by the rela;onship between images. what is most striking is the ability to rotate the shape and examine the void within the surface of the 3D composite image. This effect is not dependant on one par;cular technique or technology. Examples include MicrosoD Photosynth (2007), Adobe 123D Catch (2009), and AgisoD Photoscan (2006). Stereo-‐photogrammetry u?lises the shims in perspec?ve generated by parallax, the rela?ve posi?on of the foreground and background, to build an image surface formed around a 3D shape. The algorithm calculates the shims in rela?ve posi?on between foreground and background in order to calculate the three dimensional rela?onship between elements. It is the rela?onship between the images that generates the shape.
20
21
eg -‐ depth data generated from Google Street images h2p://www.patriciogonzalezvivo.com/2014/pointcloudcity/ ]
22
Reminiscent of Ariel Caine’s point clouds shown yesterday
23
Point cloud then connected into a structure via polygon mesh
24
25
Benne2 2015 work in progress Love the spider web effect
26
27
Hyper real
28
h2p://www.qfxdigital.co.uk/#!duologue-‐memex/c1sem
29
30
h2p://www.dailymail.co.uk/sciencetech/ar?cle-‐2761272/This-‐NOT-‐real-‐woman-‐ Meet-‐Beryl-‐creepy-‐lifelike-‐3D-‐virtual-‐model-‐using-‐scans-‐elderly-‐lady.html
31
32
33
Photogrammetry is one of a number of techniques employed to create 3D digital scans. For example, the Smithsonian x 3D, the Smithsonian Ins?tute’s 3D digitalisa?on program, uses a combina?on of techniques that includes “laser scanning, structured light scanning, and DSLR photogrammetry" [Gates 2015]. These three techniques are clearly seen in the documenta?on of the process of crea?ng a bust of President Obama [White House 2014].
34
Whilst photogrammetry is a rapidly emerging area of mainstream digital imaging – a number of phone apps are available -‐ there is a pre-‐photography historical precedent in the work of 17th century sculptor Bernini (Co2er, 2008)
35
36
Bust of Charles I, a2ributed to Jan Blommendael (Royal Collec?on, n.d.) with triple portrait of Charles I by Anthony van Dyck c1635 (Royal Collec?on, c1635) exhibited together ‘In Fine Style’ at The Queen's Gallery, Buckingham Palace (Royal Collec?on, 2013). Photo: (Bates, 2013)
37
This poten;al within photography was acknowledged at the beginning of the medium/technology. In Arago’s 1839 announcement to the Academy of Sciences in Paris on ‘a method of capturing images with a camera’, Arago noted the implica;ons of photography for the efficient collec;on of topographical data (Barger and White, 2000, p. 25). Arago viewed the daguerreotype technique as a process for scien;fic analysis, a means of mapping and measuring (Tresch, 2007). Aligns with the 19th century moment when the illustra/ons of moved away from roman/c depic/ons of ruins in a landscape to a mode of inventory and categorizing finds. (Galperina, 2014) “Equip the Egyp/an Ins/tute with two or three [examples] of Daguerre's apparatus, and before long on several of the large tablets of the celebrated work, which had its incep/on in the expedi/on to Egypt, innumerable hieroglyphics as they are in reality will replace those which now are invented or designed by approxima/on. These
38
Well before Muybridge in 1878 was using photography to deconstruct and recreate ?me through serial sets of photographs Willeme was using photography to deconstruct and recreate spacial rela?onships with 2D photographs in Paris
39
Transla?on from photo to sculpture via a pantograph h2p://en.wikipedia.org/wiki/Pantograph Note this is a USA patent
40
François Willème Unfinished Photosculpture, 1859
41
Barry X Ball has been using photogrammetry [?] to reproduce (make ‘amer’) sculptures in his Masterpieces series, including ‘Hermaphroditus asleep’, which includes a ma2ress carved by Bernini (Louvre, n.d.).
42
43
44
45
Clement Valla has also used photogrammetry to photograph museum objects for his 2014 exhibi?on Surface Survey, but his project has been an examina?on of the structures and surfaces created by the technology itself. In an interview published in Animal New York, Valla explained that he was invited to work on a project with the Metropolitan Museum Media Lab, where he discovered their 3D models. I began taking them apart to see how they had been constructed, deconstruc/ng them into texture maps produced by the soSware. They immediately reminded me of archaeological fragments, bits and shards of ar/facts to be reassembled into a complete whole… and of archaeological illustra/ons from the late 19th century, at the moment when the illustra/ons moved away from roman/c depic/ons of ruins in a landscape to a mode of inventory and categorizing finds. (Galperina, 2014) Valla draws an analogy between the pieces of texture created from deconstruc?ng the 3D photogrammetry files and the fragments found on an archeology dig. He then compares this with the possibili?es of archaeology in the digital archive. He explored this by exhibi?ng the texture maps of found 3D scans alongside unwrapped textures of the museum object scans (Galperina, 2014; Pangburn, 2014; Transfer Gallery,
46
Project Mosul has actually used screenshots from the footage of the objects destruc?on in the photogrammetry reconstruc?on So, to summarise my point, photogrammetry is one example of how new surfaces are constructed in the rela;onship between images. In this case, the surface is a composite, a shell comprised of shards that have been compiled by algorithmic computer vision construc;ng the shape based on measurement of the rela;onship between image elements and extrapola;on based on a projec;on of perspec;ve lens representa;on. The surface is shaped from the rela;onships between photographs. The digital artefact is a hollow shell that can be deconstructed into its cons;tuent parts.
47
In his 2007 TED Talk, Blaise Agüera y Arcas demonstrated how Photosynth can execute a form of photogrammetry to create a 3D model of Notre Dame Cathedral That was constructed from a collec?on of images scraped from Flickr (Agüera y Arcas, 2007). He pointed to a poten;al future where all images can be connected spacially via their visual content. This demonstra;on of Photosynth is a concrete example of the connec;on between computer vision, photogrammetry and the use of search engines to group and organise knowledge.
48
Whilst photographs have always been arranged and grouped, shared, touched and handled, the affordances of the digital environment have altered the forms of the encounter and the rela;onship between images. Most web-‐based photographs are hyperlinked to further content. The viewer is urged to click through to the next image encounter. Photographic images are experienced as a cascade of linked and interconnected image planes. The encounter is a click or stroke, flow or swipe. This rapid growth in the circula?on of photographic images can be compared to the rapid penetra?on of photography in the 19th century in the decades following the announcement of the daguerreotype in 1839. Within months of Daguerre announcing his technique in Paris, Samuel Morse had obtained a transla?on of Daguerre’s manual and established a photographic studio in the USA (“Divine perfec?on,” 1999).
49
Natale draws a correla?on between the introduc?on of photography to the USA in the mid 19th century and the growth of communica?on media – telegraphy, railroads and the postal system – that is familiar to us in terms of the contrac?ons of ?me, space and knowledge wrought by digital communica?on media in the last two decades (Natale, 2012). The figure of Samuel Morse is just one example of the entanglement of communica?on and imaging technologies. One implica?on of this current phase is that digital photographs are no longer contained as stand-‐alone two dimensional prints to be contemplated and considered. Whilst photographs have always been arranged and grouped, shared, touched and handled, the affordances of the digital environment have altered the forms of the encounter and the rela?onship between images. Most web-‐based photographs are hyperlinked to further content. The viewer is urged to click through to the next image encounter. Photographic images are experienced as a cascade of linked and interconnected image planes. The encounter is a click or stroke, flow or swipe. boundaries of photography -‐ disolved and ubitqitous]
50
In 2011, Google introduced the ‘search by image’ feature (Chris Crum, 2011). This facility differed from the established version of Google Images where searches were text based search terms. Search by Images is based on the image file itself Google ‘search by image’ is not the first reverse image search engine, TinEye was launched in 2008 (TinEye, 2008) claims to be “the first image search engine on the web to use image iden?fica?on technology rather than keywords, metadata or watermarks” (TinEye, n.d.).
51
Samuel Bland’s Googlology series is a visual conceptual strategy that reverse engineers and reveals something of the workings of the Google ‘search by image’ algorithm. Using his original photographs, Bland combined the first twelve ‘visually similar’ results from a Google ‘search by image’ search (Schiller, 2013). The search results were layered to create a composite average that reveals the workings of the computer vision algorithm.
52
Bland’s composites clearly illustrate that, at this ;me, the func;on of computer vision within this algorithm does not comprehend content or representa;on. Whereas tradi;onal organiza;onal taxonomies might arrange images according to their content, by what they represent (images of birds in one group; images of cars in another group), the Google ‘search by image’ results connect and group images according to the formal arrangement of shape and line, contrast and colour Content blind
53
This is somewhat like rearranging all the books in a library according to their size rather than their subject maNer.
54
Bland’s work is a concrete example of the challenge to representa;on raised by Rubinstein & Fisher (Rubinstein and Fisher, 2013, p. 9). In this current ‘reverse image search engine’ algorithmic environment, digital photographs may no longer sorted, organized, associated and linked according to their representa;onal content.
55
In an interview recorded in 1988 at the European Media Art Fes?val in Osnabruck, Flusser described the geometry of an emerging paradigm as ‘structural’ thinking and predicted that the implica?ons would be as significant as the introduc?on of wri?ng
56
A2empts to visualize the geometry of the internet certainly reinforces this idea of computers as a medium genera?ng complex networked structures (Meeks, 2011b).
57
I am proposing that the rela;onships being drawn between images by search engine Might also be visualised as a geometric structure rather than Bland’s flaNened composites
58
If I were to a2empt to summarize Flusser’s model, we could say that the technology of wri?ng is uni-‐dimensional and enforces/facilitates linear thinking; photographs, are two dimensional and facilitate ‘scanning’ planier forms of thinking; whereas computer technologies support a ‘structural’ three dimensional complex form of thought. I am proposing that the rela;onships being drawn between images by search engine Might also be visualised as a geometric structure
59
Flusser is by no means the first and only writer to consider the implica?ons of technology on culture, knowledge and thought. For example, Ong’s Orality and Literacy considered the shim in cultural consciousness between oral culture and the impact of wri?ng as a technology (Ong, 1982); Andy Clarke, author of Natural Born Cyborgs, discussed extended mind theory and the ways in which we use technologies to extend and facilitate thought (Clark, 2003). Observa?ons on the rela?onship between technology and thought is reiterated and extended by Rowlands with his explora?on of thought as embodied, embedded, enacted, and extended (Rowlands, 2010) . The no?on that technologies have profound impacts on consciousness are supported by a number of studies
60
We know that technologies are not value neutral. Take for example the bias embedded in photography towards causican skin (Roth, 2009; “Teaching The Camera To See My Skin,” n.d.). Ted Striphas employs the term “algorithmic culture,” as “the ways in which computers, running complex mathema;cal formulae, engage in what’s oDen considered to be the tradi;onal work of culture: the sor;ng, classifying, and hierarchizing of people, places, objects, and ideas” (Granieri, 2014).
61
Given the emergence of sor;ng algorithms for images that do not rely on representa;onal content, we find ourselves at a moment where the taxonomy of image mediated knowledge/culture has been shiDed in a profound way – from representa;onal content to formal visual elements. If we extend Flusser’s conten?on that technologies impose shape and structure on thought/knowledge to consider the implica?ons computer vision mediated search by image engines, how might we begin to conceive of how the organizing of images via computer vision will structure, shape and facilitate ways of knowing and perceiving. -‐-‐-‐-‐-‐ Mee?ng Notes (6/06/2015 11:34) -‐-‐-‐-‐-‐ poten?al to
62
This moment I have described in the rela?onship between digital photographs, computer vision and reverse image search engines may be flee?ng, if not already past. It may be no more than a snapshot in a rapid journey. But snapshots can help us to understand where we have been and where we are going, a surface ground on which to briefly rest and reflect.
63
The obvious and immediate excep?on to the content blindness of computer vision is the prevalence of facial recogni?on algorithms. Facebook’s DeepFace program claims to have an accurately comparable to human cogni?on This may be a passing phase given the emergence of deep learning AI capable of associa?ng language with image content, but it does s?ll give us a moment of insight into the implica?ons of the shim towards a seman?c web. Will we no?ce the paradigm shim? Given the connec?on with emo?onal desires met by smart phones and social media, we may not no?ce as we pass through the wormhole.
64
S?ll blind to a certain extent There is a plethora of art projects that a2empt to reverse engineer, reveal and subvert this feature of big brother surveillance. Among my favourite examples are Zach Blas’ Face Cages (Andrew Lasane, 2014; Blas, 2013) and Onforma?ve Studio’s collabora?on with Chris?an Loclair Google Faces (onforma?ve studio, 2013). These projects point to an awareness that as we look at computers, computers are looking at us, the primary concern being that computer vision has significant implica?ons for privacy. In the case of Bas’ Face Cages, the shape of the biometric algorithm
65
There is some evidence that the opera?on of computer vision may be gaining a form of cogni?on Not that we can really know if an AI has ‘cogni?on’ in an organic sense but there are increasing examples of automa?ng recogni?on. In 2012, Google Research set an ar?ficial intelligence deep learning program the task of looking at Youtube videos in order to learn to recognize content. Of course, the first thing that it learnt to recognize was cats (Clark, 2012)
66
In 2014 Microsom Research reported on development of AutoCap?on, an app to prompt users to cap?on their images by sugges?ng descrip?ons of the content (Ramnath et al., 2014). Similar to theories of human percep?on, AI deep learning protocols develop and refine from exposure to content (Wolchover, 2014).
67
More than matching words with text, recent work at UC Berkley has been mapping brain ac?vity with known mind’s eye visuals. In 2011, a group of researchers demonstrated some early results in work to literally capture images from the mind’s eye using MRI mapping (Anwar, 2011; Nishimoto et al., 2011). In her Moonshots presenta?on and TED Talk, Mary Lou Jepsen discussed the achievability of this technology to deliver usable results (Solve for X, 2012; Jepsen, 2013). She also introduced research that suggests that it may be possible to get the receptor neurons in the re?na to run in reverse and literally read images from the mind’s eye from the eye itself. This proposi?on has interes?ng resonance with the historical concep?on of vision as an emission rather than an intromission. Curiously, this debate may also have some connec?on with phenomenological concep?ons of vision. Sadly, I do not have the space to inves?gate this in depth at this point other than to also men?on that a phenomenological analysis of the no?on that ‘the computer looks back’ may be rich and to note that Wiesing (2010) has already considered the presence of phenomenological thinking in Flusser’s work. Tesla’s 1933 proposi?on for a ‘thought camera’ may not be so kooky amer all.
68
Tesla’s 1933 proposi?on for a ‘thought camera’ [reference] may not be so kooky amer all. 1933 Deseret News -‐-‐-‐-‐-‐ Mee?ng Notes (6/06/2015 14:37) -‐-‐-‐-‐-‐ light field AR/VR h2p://placefacecyberspace.net/2015/03/05/thought-‐camera/
69
To summarise: Technologies create paradigms computer vision is media?ng the emergence of forms of surface structure between digital photographs
70
I am not a2emp?ng to predict precisely what the implica?ons will be But no?ng that computer vision mediated rela?onships between images has the poten?al to reshape the structure of thought. I propose that MESH is a fer?le metaphor with which to consider the shape of this emerging algorithmic image environment It describes the way in which surfaces are built between images – both photogrammatry and reverse image search engines It also describes the embeddedness, the enmeshment of these structures in our culture. -‐-‐-‐-‐-‐ Mee?ng Notes (6/06/2015 11:34) -‐-‐-‐-‐-‐ it also points to the complexity of the structure and surfaces generated by an algorthimic photographic environment the surface is an approxima?on, a sugges?on, that become porous as we approach
71
Natale, S., 2012. Photography and Communica?on Media in the Nineteenth Century. Hist. Photogr. 36. Na?onal Gallery, n.d. Philippe de Champaigne and studio | Triple Portrait of Cardinal de Richelieu | NG798 | The Na?onal Gallery, London [WWW Document]. Na?onalGallery.org.uk. URL h2p://www.na?onalgallery.org.uk/pain?ngs/philippe-‐de-‐ champaigne-‐and-‐studio-‐triple-‐portrait-‐of-‐cardinal-‐de-‐richelieu (accessed 1.2.15). Nishimoto, S., Vu, A.T., Naselaris, T., Benjamini, Y., Yu, B., Gallant, J.L., 2011. Reconstruc?ng Visual Experiences from Brain Ac?vity Evoked by Natural Movies. Curr. Biol. 1641. doi:10.1016/j.cub.2011.08.031 onforma?ve studio, 2013. Google Faces. Ong, W.J., 1982. Orality and literacy : the technologizing of the word, New accents. London ; New York : Methuen, 1982. Pangburn, D., 2014. Become A Digital Archaeologist With Clement Valla’s “Surface Survey” Exhibi?on [WWW Document]. Creat. Proj. URL h2ps://thecreatorsproject.vice.com/blog/become-‐a-‐digital-‐archaeologist-‐with-‐clement-‐vallas-‐surface-‐survey-‐ exhibi?on (accessed 12.28.14). Payne, M., n.d. Mimobase [WWW Document]. h2p://mimobase.com/. URL h2p://mimobase.com/ (accessed 1.4.15). Photoshop CS3: Final release details [WWW Document], 2007. . Digit. Photogr. Rev. URL h2p://www.dpreview.com/news/ 2007/3/27/pscs3 (accessed 5.31.14). Prosthe?c Knowledge, 2014. 3D Printed ArcheAge [WWW Document]. Prosthet. Knowl. URL h2p:// prosthe?cknowledge.tumblr.com/post/98012948036/3d-‐printed-‐archeage-‐curious-‐li2le-‐project-‐by (accessed 3.16.15). Ramnath, K., Baker, S., Vanderwende, L., El-‐Saban, M., Sinha, S., 2014. AutoCap?on: Automa?c Cap?on Genera?on for Personal Photos. Microsom Research. Roth, L., 2009. Looking at Shirley, the Ul?mate Norm: Colour Balance, Image Technologies, and Cogni?ve Equity. Can. J. Commun. 34, 111–136. Rowlands, M., 2010. The new science of the mind : from extended mind to embodied phenomenology. Cambridge, Mass. : MIT Press, c2010. Royal Collec?on, 2013. In Fine Style: The Art of Tudor and Stuart Fashion [WWW Document]. R. Collect. URL h2p:// www.royalcollec?on.org.uk/exhibi?ons/in-‐fine-‐style-‐the-‐art-‐of-‐tudor-‐and-‐stuart-‐fashion-‐0 (accessed 3.16.15). Royal Collec?on, c1635. Sir Anthony van Dyck (1599-‐1641) -‐ Charles I (1600-‐1649) [WWW Document]. RoyalCollec?on.org.uk. URL h2p://www.royalcollec?on.org.uk/collec?on/404420/charles-‐i-‐1600-‐1649 (accessed 1.2.15). Royal Collec?on, n.d. A2ributed to Jan Blommendael (c.1650-‐1699) -‐ Charles I (1600-‐1649) [WWW Document]. RoyalCollec?on.org.uk. URL h2p://www.royalcollec?on.org.uk/collec?on/35856/charles-‐i-‐1600-‐1649 (accessed 1.2.15). Rubinstein, D., Fisher, A., 2013. Introduc?on: On the Verge of Photography, in: On the Verge of Photography: Imaging Beyond Representa?on. ARTicle Press, Birmingham. Schiller, J., 2013. Google Is Alive, It Has Eyes, and This Is What It Sees | Raw File [WWW Document]. WIRED. URL h2p:// www.wired.com/2013/05/sam-‐bland-‐google-‐goggles/ (accessed 5.31.14).
72
