Refik Anadol’s animations – meaningful moving parts recombine to generate art

Posted on by
Reply

Refik Anadol has made his name by producing super scale animations that rely on machine learning algorithms, a large cast of studio assistants, and meanings which are borrowed from the raw material consumed by the algorithms. His invited exhibition at the Serpentine in London, entitled Echoes of the Earth: Living Archive, takes over the whole of the Serpentine North Gallery, is totally eye-catching. The TL;DR is: go. Quickly. It closes April 7 2024.

The images in the exhibit range from literal-seeming to highly abstract, and the animations which sequence them flow from one mode to the other. Everything is synthetic – the images are derived from applying machine learning algorithms to very large datasets of other, real images. On which more later. The results sometimes seem literal – as with this synthetic coral:

Coral-like image
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

Sometimes the images produced by the operation of the models are highly abstract, as in this candy coloured image, which seems fluid-like and tissue-like, but does not obviously represent any particular thing:

Abstract multi-coloured particle-based image
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

The animations often looks like they are driven by a fluid dynamics simulation. They proceed at a comfortable pace. The colour palettes are harmonious, the surface lighting is gentle. Because of the palette and palette transitions chosen, the sequences have a tendency to look “nice”.

The visuals aren’t always explicitly representational, although the models which generate them use real image captures as inputs. But even when they aren’t interpretable as specific things, the outputs all have a biomorphic flavour to them, reflecting the characteristics of their inputs. Some sequences evoke associations of clouds of organic tissues in perpetual motion and transition. As indeed real weather and real tissues are.

Harmoniously coloured abstract image sequence – reminiscent of changes in weather and in living tissues
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

Sometimes, the images flow from the purely abstract towards representational, then back off again. This gives the feeling of life-forms being created – and dissolving. The pace is slow enough that it doesn’t feel frenetic, rather graceful.

Synthetic creatures emerge from more abstract images – and dissolve again
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

As is the way with the state of the art around synthetic ML-based images of beings, there are times when the synthesis creates “errors”, as seen in this bird with two beaks, and a foot whose claw doesn’t grip the surface:

Model outputs are driven by visual similarity not physical constraints – check the beak(s) and claws
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

What are we to make of this? The images being synthesised and streamed together explore the latent space of the models – creating images of possible beings. The models are based on very very large datasets of actual images, but the model outputs are novel. Sometimes they create things which we instinctively understand would be unable to exist in reality.

In a way, these model “errors” serve to highlight the preciousness of actual life and beings. Not all corners of this very effortfully produced and highly dimensioned latent space are actually habitable. Visual representation is fundamentally a surface phenomenon, with only imperfect connections to mechanism.

Sometimes what is created is beautiful, as in this sequence of a lotus turning into a lotus:

The model dreams impossible dreams – of a lotus within a lotus
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

There is treasure to be found in mining latent spaces, and Anadol and his team have certainly found some. Dreams, even deep ones, do sometimes turn to nightmares though. The potential for monstrosity, as well as beauty and impossibility, is equally latent in the technique.

The story of how the base data images are sourced, and how the resultant images are synthesised is very much a part of the work. Which is very much on trend. This display, which is part of the exhibit, dramatises the multitudes of base data images that went into creating the models:

The origin story is part of the work
Refik Anadol Echoes of the Earth: Living Archive
Serpentine Gallery London 2024

The works are made possible by the collaboration of an impressive number and range of institutional collaborators, such as the Natural History Museum, and the Smithsonian, who have collectively contributed millions of images from their own collections to be used as training data. Many of the works use what Anadol calls their Large Nature Model, a play on Large Language Models (LLMs) of ChatGTP fame. (This model is, apparently, open sourced – although, interestingly, I am not the only one who can’t find it on GitHub.) All the works seem to use NVidia StyleGan2 as the underlying model framework.

This is the first of Studio Anadol’s installations that I’ve seen in person. In concept and style it has strong links to his blockbuster 2022-23 installation, Unsupervised, at MOMA in NYC. Unsupervised was a never-ending generative machine learning-based animation using, as its raw material, 200 existing works in MOMA’s collection. This use of freely contributed material and generative ML-driven animations are common threads tying the the two exhibits together.

The MOMA installation was accompanied by righteous chat about it being “on the blockchain” – but times change. The Serpentine exhibition also surfs the zeitgeist, but doesn’t mention crypto. Now the force being harnessed is the sharp wind of eco-anxiety. This is somewhat ironic, given the compute power – and energy – which must have been involved in creating the exhibition’s works. Anadols’ explanation of the intent and process behind the highly collective work – of which the Serpentine exhibition is just the first public manifestation – was launched at Davos last year: you can check out the promo here.

So? It’s an eyeful, in a good way, and it represents boss levels of effort and collaboration. But the outputs – although they are “nice” – make me melancholy. In exploring the latent corners of a many thousand dimensioned model space built from a multitude of images of living things – what have we learned? That actual, working life is precious in its uniqueness? It’s a bold exploration, and it’s certainly uncovered some beauty, but I don’t feel the exploration has helped us, collectively, arrive anywhere. Yet.

Getting real with AR – the magic, the mundane, and the mysterious

Posted on by
Reply

Remember AR? After years of incubation in innovation labs, it’s still pecking away at the inside of its shell. The recent Herzog & de Meuron exhibit at the RA illustrates both the potential and pitfalls of AR/MR design. Ironic, in a way, that an exhibit so wholeheartedly about design – and the importance of model-making, concept co-creation, and process in design – also serves as a powerful illustration of the challenges of successful AR design.

Herzon and de Meuron is a Swiss-headquartered global architecture firm. Londoners might know them as the designers of the Tate Modern, the 2012 Olympics site master plan, and new Chelsea F.C. stadium. Amongst other projects. You’ve seen their work even if you didn’t realise it at the time. I went to the exhibition as I’m interested in architecture, in general, and am a fan of their work, in particular. And I also love architectural models. That H&deM had developed an AR app specifically to support the exhibition was, for me, initially just an amuse-guele. But the ways in which the app succeeded and failed turned out to be very interesting: they suggest a few principles for the AR grimoire, in the way of Do’s and Don’ts.

Magic

For me, the most exciting experience in the exhibition app was the full scale overlay on a real room sized model. The 3D model surface was plain white – when viewed with your eyes only. Viewing the model surface via your phone (and via the app), you saw a set of decorative elements superimposed on the blank contours of the model, rendered with appropriate shadows.

H&dM RA exhibit 2023 app
Screen capture of AR view of room scale model Project 377
There is no wood trim – it’s AR

This model is taken from the design process of the (beautiful) Kinderspital (Children’s Hospital) in Zurich, a project that H&deM have already put over a decade’s worth of work into. Of course, full scale models aren’t always feasible in prototyping work. But when dealing with such a diversity of stakeholders – patients, researchers, staff, benefactors – I can see they could be useful in helping to reveal preferences and needs.

However, what struck me most about this work wasn’t that it was useful. Or even that it was beautiful. It was – simply – that it was magic. It was magic to see the normal world transformed, in real time, at real scale, and believably – by simply by viewing it through the lens of my phone. It’s typical that AR apps superimpose views of digital objects over your view of the real world. Dinosaurs roam football stadia. Butterflies flit in the auditorium. Pleasantly surreal, yes, but not magic.

What made this magic as an experience was the experience of a total integration of the 3d physical world and the AR overlay. I could move from side to side, and look up and down, and walk, and the overlay correctly preserved registration of the service, and the perspective I would see it from. It really seemed to be there – and part of my physical environment, in a normal way. It was the simple normality of it that I found to be magical.

Maybe not everyone finds fake wood trim so exciting. But the principle of perceived integration with the constraints of the physical environment is something that can be applied more broadly. The dinosaur in the stadium can bump its head on the stands. The butterfly can emerge from projection mapped 2D into 3D free space. Simple AR stickers that float context free simply don’t have the same persuasive power.

Can research create magic? After years (er, decades…) of doing an entertaining variety of different types of reseearch, I’d say probably not. It can certainly improve alignment between the realisiation of a service or product and the needs of its users and investors. Also, research can – in time – be used to generate, articulate and refine design heuristics that are useful in mapping out more and less promising areas of a design space, and in explaining – in hindsight – why some things work and others don’t.

The mundane: people do the darndest things

There’s also the more low level but very useful approach to design research, of just plain old seeing what people do when they are faced with a design they can interact with. A prototype design is fine – as H&deM well know, as they have based their core architectural design process on exploration. There is very little in design that doesn’t benefit from being wind tunnelled by actual ordinary people, and their actual ordinary confusions and behaviours. This holds for apps as well as architectural designs.

So, seeing people at the exhibit trying to interact with the AR app – and failing – was very instructive. It would have been great if H&dM had been able to really check out how people used their AR app before setting it loose on the world, as the issues I saw just from watching people for a half hour or so were pretty easy to fix. I’d batch all the snags into the same bin, as discovery problems, basically:

How do you make it go?
Is this all there is?

Generic discovery challenges with the H&deM AR app

How do you make it go?

AR-enabled exhibits were indicated by a non-functional icon on the exhibit card
H&deM RA Exhibit 2023

The symbol indicating the availability of an AR layer to an exhibit was an icon of a sphere in a box. I saw one gallery-goer aim his phone at the icon on the exhibit card, expecting it to activate the AR experience. Not unreasonable – in hindsight. However, the icon was just a signal that you needed to search elsewhere in the display for a QR code card. Not everyone discovered this. Separating the actionable cue (the QR code) from the informational cue (the icon) kept the information card for an exhibit looking lean, but introduced an extra step not everyone knew to take.

Is this all there is?

Once the QR code had been located, sometimes there was an extra challenge, which was to understand whether or not you were seeing what you were supposed to, and what it was. There was one exhibit where the overlay wasn’t visible until you backed up almost into the facing display case. From what I could see of how people were using the app, most people didn’t usually discover there was any there there.

There was another exhibit, where what you saw was just a label for one of the exhibits. But of course that might have meant I missed. the point – as it seemed quite pointless. There were a few more exhibits where AR “stuff” floated above the physical exhibit but its relationship to the physical layer was unclear.

What the AR app showed was sometimes quite mysterious
H&deM RA Exhibition 2023
The relationship of the AR overlay to the physical exhibit wasn’t always obvious
H&deM RA Exhibition 2023

Part of the fun of exploring an AR layer for a display of architectural models is that there aren’t any rules about what to expect to see in the layer or how to interact with the layer. Fundamentally it’s going to be a discovery exercise. But the sheer variety of options presented in the H&deM app – often without much or any cuing – made it difficult for me to know whether I was actually seeing something “as designed” – or whether I’d managed – somehow- to miss the point entirely.

These confusions about the use of the exhibit app aren’t complicated ones – they are easy to detect by observing how people interact with the app. They are also solvable in various ways. More explicit cuing off-app could be useful- and damn the torpedoes around keeping the real world graphics lean and mean. And a more limited (and therefore predictable) repertoire of options for what the AR overlays disclose would also help people onboard to the experience more easily.

C’est magnifique, mais… qu’est que c’est que ça?

There was another digitally enhanced exhibit I enjoyed just as much as the room scale fake wood trim. I’m not sure if you’d call it AR. It probably isn’t. But it’s lovely. It’s an exhibition-wall-scale diagram of a floor plan for the Kinderspital in Zurich. I’ve never seen a plan that big. For massive projects, on a plan this big you can see features that would be invisible at other scales. I’m intrigued to think of how this might help people understand how the design works, or how to build it efficiently. (I’m not actually sure what the answer might be. Needs a bit more thinking.)

Not only is it impressively massive, but it has projection mapping overlays that change and pick out and highlight different aspects of the design. I found it interesting even without the overlays. But the overlays are certainly a delightful icing.

Grand scale visualisation of a grand scale project
H&deM RA Exhibition 2023
Multiple views highlighting different aspects of the plan are delivered via projection mapping
H&deM RA Exhibition 2023

So it’s magnificent, oui oui, but is it AR? Yes, in a way, in that it’s a combination of physical – in this case, a giant scale plan of a giant building – and digital augmentation, delivered through the various projection mapping overlays that are possible. No, in a way, as it seems like pretty straightforward static data visualisation with multiple viewing options – albeit at a truly heroic scale. One element is physical and the other digital, so it is “digitally augmented”, but the same effect could be accomplished – less heroically – with an overhead projector. I think part of my feeling that it’s not “real” AR is that it doesn’t do anything. But architectural models typically don’t, on their own, do anything. Their usefulness comes from their ability to support thinking about them, and simulating experience. The interactions it supports are open ended, and non specific. However the potential now exists to do more, and have the model participate more actively in interactions with it. With advances in adoption and capabilities of BIM (building information modelling), I think we can look forward to more and even better interactive capabilities where this came from in future.

In summary

I was expecting to find the exhibition delightful. I like architectural models and I like H&deM’s work. And indeed so I did. I wasn’t expecting the exhibition to have an AR app – or for the AR app to provide such a lot of food for thought about what works and what doesn’t. So, for me, the fact the AR view of the exhibition contained both moments of delight and things that really needed fixing was a bonus.

The RA Summer Exhibition 2023

Posted on by
Reply

I’ve seen the RA Summer Exhibition in different lights over the years. Once, long ago, I went to an opening night sponsored by my employer of the moment, whose shiny offices in Berkeley Square were just a few streets away. Everyone was busy hustling and necking drinks from trays, and nobody was looking at the exhibition, or even talking about it without having looked at it. It seemed almost rude to attend to the art instead of the event. Nowadays I know better than to go along with hundreds of others who – despite being there – are intent on not attending.

Convergent evolution

The exhibition has been running annually since 1769. In recent decades, I’ve seen it change in interesting ways. For one thing, it attracts a different crowd. Year by year, I’ve been more and more able to blend in. Partly this is because I’ve been getting older. But also the crowd is less ironed, more relaxed, and less social x-ray than it used to be. Cool would be going too far. But the audience vibe is now a blend of Goldsmith’s degree show (half price student tickets, and the national art pass have helped here, I would guess), earnest tourists, and suburban culture vultures. Only just the faintest whiff of Met Gala.

What’s on show in the way of art, as well as audience, has also changed:

  • Curation is more varied and attribution is more transparent – with interesting consequences
  • There are some disturbing ‘concept’ pieces and digital works – for better or worse

Also, there are some things that haven’t changed in the past several decades, or perhaps the past few hundred years:

  • Pets, seascapes, and clever but gentle paradoxes seem to do well – for the record, in case our feline overlords are reading this, I’m not against that in the least
  • There’s still a wild jumble sale feel to it all – as you’d expect from an exhibition with over 1,614 objects on display.

Connections, curators, and consensus

This year there is a theme, Only Connect, but you’d be hard pressed to guess it from looking at the works. It’s so universal as to be almost content free. There is an overall exhibition coordinator, David Remfry RA, but quite wisely he has not tried to boil the ocean all on his own. There are eight other committee members, all RA artists. Each room of the exhibition is curated by a single named individual. This makes for more coherent presentation, as each room reflects each curator’s eye.

Two works, in particular, caught my attention from a curatorial point of view. Eileen Cooper RA picked out a work from Sir Richard Long’s The Tide is High series, Tideline, shown in Gallery 8.

909 – TIDELINE, FROM: THE TIDE IS HIGH
Sir Richard Long, RA

Richard Jacklin RA also picked a work from Sir Richard Long, RA from the The Tide Is High. I think you’ll agree with me that from certain angles it looks similar.

HALF TIDE, FROM: THE TIDE IS HIGH
by Sir Richard Long, RA
No. 444 – RA Summer Exhibition 2023

What’s interesting to me is that it’s the only overlap between each of the rooms. The RA gets about 20,000 submissions for the summer show each year. They accept over 1,000, but under 2,000. Isn’t it interesting that only one work – actually a near-duplicate, not a completely literal duplicate – made it into two curator’s rooms? Many of the works are prints, so it wouldn’t require a rent in the fabric of of space time, or anything like that, to make it happen.

If curator picks happened purely by chance, you’d expect more than one item of overlap in the nine curators’ picks. Similarly, if there was any kind of barometer or lowest common denominator of “taste” that the curators shared, you’d also expect there would be more than one piece of art caught in more than one net. But that’s not how it falls out. Given any individual curator, what they curate assorts to be maximally distant from what anyone else curates, like the output of a well fitting cluster analysis.

It follows that any work that any two curators agree on is likely to be worth a look. And in fact I like both these excerpts from Long’s The Tide is High series very much. The artist, Sir Richard Long, is known as a conceptual artist – he made his name in the 1960’s photographing a path he forged out in grass. But it has a spare elegance which reminds me both of classical Japanese decorative art, and modern generative art. It’s also grounded – in a very specific and literal way: it is part of a series devoted to documenting and re-interpreting patterns in tidal mud. This is a lovely tension. To me, the work is timeless and classic.

Nods to the new

I remember much work at the RA Summer Exhibition in times past being pleasant and traditional. Overly so. But there are now more works which bite, or are otherwise difficult. One is the kinetic sculpture from Giles Walker (one of the founders of Mutoid Waste Project). This animatronic work shows a robot child – hooded as if being tortured, with wires sticking out of its belly, standing on a low chair and jerking in an uncanny way. The work is called It takes a village to raise a child.

It takes a village to raise a child, by Giles Walker
RA Summer Exhibition 2023

Giles Walker says, in the interview on the artists interview show reel for the exhibition:

“considering we’ve got a million kids living in poverty I guess that says a lot about our community as a whole”

Giles Walker, RA Summer exhibition Artist Interview show reel

So, with the video, the animatronic can be seen as a work of social criticism about our treatment. of children. Maybe it is also intended to contrast Hillary Clinton’s book, It take a Village, with the use of torture in interrogation by the same society.

Without the accompanying video, and without knowing the title – titles are not displayed alongside any of the artworks at the the show – the work fully succeeds in being disturbing, but does not have a specific interpretation. Although it’s clear there’s something wrong, it’s not clear what it is.

Bog Man, a sculpture by Carlos Zapata, also has an uneasy feel to it. It’s a life sized piece that probably does what it says on the tin – depicts the mummified corpse of a man who died in a bog. It’s not clear which man. The twisting of the body suggests agony – more like the remains found in Vesuvius – and the posture, powerlessness. It’s as though he is trapped in his final moment of suffering. I’m not sure what to take away from it – let alone how it relates to the theme Only Connect. Unless it’s about Bog Many connecting with his death. Or connecting with the murderous hands of his fellow creatures, as many of the bog bodies in archeology seem to have done. Whatever the specifics, I found it disturbing. Which is something.

Bog Man, by Carlos Zapata
Royal Academy Summer Exhibition 2023

Also disturbing, physical, and death-related – but with added “conceptual” – is Grenfell Tower – The Sleep of Reason by Paul Benney. The tower itself is central to the painting, looming darkly and ghostly in the composition. The conceptual twist is that the image itself is made from debris of the defective building cladding from the disaster which killed 72 residents. Does this make it more ‘real’ in some way?

Grenfell Tower, the Sleep of Reason, by Paul Benney
Royal Academy Summer Exhibition 2023

Clearly the Summer Exhibition isn’t just for niceness any more. And the media choice is broadening – slowly – as well. There were architectural models (always a win), and a few digital works. I’m an aficionado of digital, as a creative medium, but I’m curious about whether artists working in digital are actually submitting their works to the show in sufficient quantities to give the curators a really wide and deep choice of selections.

Cats, seascapes, and cosy mysteries rule the dots

Most of the works on display are for sale. This adds a new dimension to the experience beyond trying to second guess the intent of the curators, the artists, and spectators. Print editions are particularly interesting as a single work can attract multiple dots, like an idea in a workshop. It’s always interesting to see which works actually move people to get their wallets out. I attended in late June and at that point I spotted three red dot patterns:

  • works with cats and other animals
  • water scenes or seascapes
  • mildly paradoxical clever concepts.

There were pleasant works in all of these categories; some were interesting. Two of the conceptual/paradoxical works that attracted a good level of dotting particularly so.

Green light – by Cornelia Parker RA is a print depicting a print that’s sold a lot and has a lot red dots on it. As of August 15th it had sold 30 of its edition of 100. I gave this piece a smile when I passed it – but only realised this piece was by Cornelia Parker when I looked it up just now. It is an interesting feature of the exhibition’s tradition that works on display are only identified by numbers, and you have to cross reference to the catalogue to find the title and the artist. I have been a Cornelia Parker fan for decades – ever since I suspected her of using our garage for inspiration for Cold Dark Matter. So finding out that this was was her work added to my enjoyment, even though my initial smile was purer.

Call by Keith Bernstein is a black and white print with a sold out edition of 40 that looks ever so much like a black and white version of his colour print entitled Call Me, distributed by Saatchi Art. What I like about both works is the contrast between the two spaces – the infinitely open Mojave desert, and the tiny phone box, lit up against the fading light in the big sky. Inside the phone box, there is a couple dressed in city clothes, catching a smooch and a clinch. The figures have an Edward Hopper Nighthawks vibe to them, artificial lighting and all. The paradox in this piece, for me, has to do with the way the phone booth seems entirely unnecessary – it’s not raining and they aren’t making a call. Surely they could equally as well be smooching in the great outdoors. But as a scenario prop the phone box works hard, by framing the lovers in a concentrated and lit space that gains meaning by virtue of the contrast with its surroundings. That the phone box seems out of place brings a weird energy to the scene that it wouldn’t have if the phone box was, say, on a street corner. Or perhaps there is no mobile signal in the desert, and, unlike a city phone box it’s actually functional. I suspect it’s meant to be archetypal and not taken too literally.

Call, by Keith Bernstein
No. 428 – RA Summer Exhibition 2023

Some pieces with good sales managed to tap into more than one “super theme”. One example is from Newton Blades, a duo as of August 15th have sold over a quarter of their edition of 100, with a combination of seascape and clever concept in Seascape N. Framing the stormy seascape as it might look from a pair of binoculars isn’t original – it puts me in mind of low tech mid-century film effects, i.e. distancing in a cosy way. But it also unavoidably placed me, the viewer, in a state of mild peril, by forcing the perspective of someone who is probably rocking from side to side a fair amount in an angry ocean.

Seascape N, by Newton Blades
Royal Academy Summer Exhibition 2023

Next year I look forward to seeing works which combines not just two but three superthemes. Although cats at sea is perhaps not the most obvious combination of themes.

The joy of the jumble sale

I know a number of people who used to attend but don’t any more, as they find the experience too chaotic and not rewarding enough. I think you have to really love the thrill of of the chase.

I’ve talked about just six works here, and there are over 1600 in the show. For better or worse. I am sure that if you went, you’d pick out six different works yet again to talk about.

I think that’s the fun of it. Although the exhibit is curated, and the curators definitely do a lot of work sifting down the 20,000+ entries the event attracts, what gets through is still very varied. Perhaps if the curators were asked to select just eight works, we would see tighter theming, more coherence and even more predictability. What this would sacrifice, though, is the joy of being your own curator, at scale. It is a jumble sale – and needing to hunt to find treasures is part of the joy of the experience.