by Glenn Wustlich
Oh hi there! 🙂
Welcome to my Visual Research document! Here I will tell you about my journey and the obstacles I had to overcome to realize my concept.
First, let me give you a proper introduction.
What is Visual Research?
Visual Research is about the knowledge and experience of design research methods which use image as source and as a tool to gain insights and distilate the user context. On one hand the Images -from observations and/or cultural artifacts- get analyzed to better understand the context. On the other hand vizualisations used as a tool to order, combine and concretize insights in the vorm of models, diagrams and visual metaphors also known as mapping.
I have no idea what this all means but it’s a literal translation of the course description so here ya go.
Henry gave us a lecture via Skype which was very interesting! He told us about several different snakes and how their patterns and colors correlate to their lifestyle and locomotion. Slow and heavy snakes tend to be camouflaged while fast moving, slender snakes are often striped or unicolored.
“The predominant hypothesis is that striping and unicolor patterns interact with motion, making it hard for predators to accurately grab the snake.
Because the whole snake is the same color, it’s hard to just where on the snake’s body you’re looking at, so when you reach for it, your hand gets there too late and the tail has already gone by.”
The goal is to collect data about the human perception on snakes, with their different colors and patterns in relation to different backgrounds and movement. Henry gave us a couple of implementation ideas to use as inspiration. Such as a hybrid of a endless-runner and a tapping game like Temple Run X Ant Smasher, where you have to tap the snake to catch it and have to avoid potential poisonous snakes.
- At high movement speeds, stripes will make it more likely for people to miss
- At high movement speeds, people will be unable to distinguish mimics from toxic species
- Camouflaged snakes will be undetected when still more often than striped, unicolor, or ringed snakes
- Cover will accentuate these differences
Visual Research Rourse Requirements
1. Vijf belangrijkste schermen incl. introductie van de app.
2. Gebruikersdata: leeftijd, man/vrouw, en beroep.
3. Statische versus bewegende patronen met verschillende achtergronden.
4. Beweging van slangen patronen.
a. Hoeveelheid vervaging
b. Soorten vormen; vierkant, cirkel, rechthoek liggend. c. Hoeveelheid vormen
a. Minimaal drie: Natuur, Stad, …
b. Dag en nacht.
c. Hoog contrast en bijna geen contrast in kleur.
6. Verschillende hoeveelheden patronen op één scherm: 2, 4, 8, 16 7. Verschillende snelheden: 15 km per uur?
8. Basiskleuren 11: zwart, wit, grijs, bruin, blauw, groen, rood, geel, roze, paars en oranje.
(Bron: Allen, W. Baddely, R. Scott-Samuel, N. Cuthill, I. (2013) T he evolution and function of pattern diversity in snakes. Behavioral Ecology.)
9. Basisvormen 3: vierkant, cirkel, rechthoek liggend.
10. Basispatronen 9: stippen positief en negatief, strepen verticaal en horizontaal, vlekken.
Civiel ingenieur bij een overheidsorganisatie
“Ik ga ervan uit dat de gegevens die ik verzamel nuttig zijn en bijdragen aan de grotere wetenschappelijke onderneming.”
Rick Moormans is een man van middelbare leeftijd met een diploma in civiele techniek. Hij is al jaren een enthousiaste buitenmens, natuurfotograaf en werkt graag in zijn tuin. Hij houdt van fotograferen tijdens zijn natuurwandelingen, die meestal plaatsvinden in natuurparken maar af en toe op gezinsvakanties. Mede hierdoor is hij zeer geïnteresseerd in lokale en inheemse flora. Rick heeft jarenlang Natuurmonumenten ondersteund, zowel financieel en als deelnemer aan begeleide wandelingen. Bovendien heeft hij zich sinds vorig jaar beschikbaar gesteld als vrijwilliger voor managementactiviteiten. Hij is geen computer- of gegevens deskundige, maar voelt zich prima thuis met computers en internet. Door zijn werk is hij bekend met het volgen van gestandaardiseerde protocollen.
Doelen en motivatie
Rick ziet zijn ‘burgers wetenschappelijke’ activiteiten als een van zijn hobby’s. Dat betekent echter niet dat hij dit werk niet serieus neemt. Hij is van mening dat de gegevens die hij helpt te verzamelen belangrijk zijn, zowel voor de behoeften van de wetenschap, maar ook als onderdeel van de grotere hoeveelheid bewijsmateriaal met betrekking tot de rol van natuurgebieden en de bescherming van diversiteit in de natuur. Zijn gevoel voor de exacte rol van zijn bijdragen is echter vaag, en hij had graag een beter idee van hoe de informatie wordt gebruikt, en ook hoe belangrijk zijn specifieke bijdragen is voor een onderzoeksproject. Hoe dan ook, omdat hij deelneemt aan het verzamelen van gegevens voor de lol, verwacht hij te blijven deelnemen zolang er mogelijkheden zijn om dit te doen.
Gebaseerd op: https://www.dataone.org/personas/rick-citizen-scientist
SNAKES ON A PLANE
“I’ve had it! With these m*therf*cking snakes on this m*therf*cking plane!”
Initially, instead of an app for smartphones, my idea was to use Virtual Reality (VR) as the medium for this citizen science project. It will yield more accurate and valuable data because we can target many of our senses in a more and precise way, which can be comparable to real-life situations. This is because we can make use of depth, vision deadzones, and a full 360 view. It is also very modular as we could implement 3D sound on a later stage for other research purposes.
I will loosely follow the requirements listed above as these where mostly targeted at smartphone apps.
I want to make an environment in which participants can do certain objectives. For example: There is one snake in the environment and the user got the task to find it. Another example could be that yet again, the user has to find a snake but in this case there are multiple snakes in one environment. We can extract data like which snake is often found first etc. The participants can complete this project at their own pace as the game is segmented in levels. Everything will be instructed by an Artificial Intelligence and the interaction methods will be as natural as possible. So there will be no difficulty curve to learn. As an extra, the snakes that the player catches, will be added to a in-game catalogue! Here, they can read more about the caught snakes and find more information. Can you find and catch them all?
Potential (future) features
First, I did some research on what kind of environment I wanted to make and use. The woods is the first thing that came to my mind. It’s lush and peaceful. It is welcoming as that in summer comes next! And who doesn’t like summer? Like in real life, the density of vegetation will also make it more challenging to find the snake. I wanted to go with an autumn setting as I think that the warm colors will give the participants a more welcome feeling. Also I do find the colors very pretty!
As I want the data to be accurately mirror real-life results, it has to be important for the environment to be convincing by replicating many real-world elements like the density. Although, since the hardware power of my Oculus Go is limited, it’s a rather difficult technical and artistic challenge.
A link to the past
The Legend of Zelda is one of my all-time game favorite game series, by one of my favorite game companies: Nintendo. Compared to Microsoft’s Xbox and Sony’s Playstation, Nintendo always had the weakest hardware in terms of graphic power. This didn’t limit them to make some of the greatest game in existence! The Zelda games where often broad and ambitious and required the developers to get the maximum out of the console. Instead of falling victim to the weaker hardware, they embraced it and came up with clever, creative ideas to realize their vision.
When thinking about solving my technical challenge, The Legend of Zelda: Skyward Sword sprung to my mind. The main setting in that game consists of floating islands with gorgeous vistas and viewing distances. Normally this would be very taxing, since the hardware has to draw all this geometry like buildings and mountains for the player to see. However, instead of doing this, the developers made a cool artistic choice which would also benefit the technical limitations. They implemented a filter which kicks in on distant objects. This filter gives a ‘painting’ effect to the scenery which gives the impression of looking at a digital dynamic painting! This way they can leave out many details on the objects whilst still maintaining the illusion of things being ‘there’. The player still sees the overal shape and color and their brain can figure the rest out. This idea of using a (oil) paint or Kuwahara like filter could benefit my prototype and might be aesthetically pleasing.
Screenshot of Skyward Sword
First attempt filter in 2D
The Gestalt Principles
Not to be mistaken by the Gestapo Principles
With my previous filter attempt in shambles I had to look at other methods to make my environment. I’ve decided to focus myself entirely on using the Gestalt Principles (“Gestalt” is German for “unified hold”). These are a set of laws from 1920s’-ish psychology. It describes how humans perceive objects by seeing structure, logic and patterns. It helps us makes sense of our surrounding world. I can make use of these tricks to let people perceive complex elements by using simple patterns with simple objects. This is also the lessen the amount of stimuli in the environment, to keep the data as pure as possible. I will further explain some of these principles I’ve used in the chapters down below. However, if you need some more explanation about these principles, you can check them out here and here.
Laying it out
First I made a 2D draft of how I wanted the scene to be. I decided to keep the play area small so that our data will be focussed to just within this space. Besides more different factors like: how good the eyesight of the subject is will come at play with larger environments, thus resulting in cluttered and more complicated data.
The scene needs to be coherent and consistent in terms of style and aesthetics. Therefore it is important to know how my objects, like the snake are going to blend with the scenery. I want to focus on colors and basic shapes. I’ve been using the autumn-ish colors picked from the Kuwahara image I made. Also, for less important elements I make use of 2D sprites instead of 3D models. Once again this is to reduce the amount of unwanted stimuli in the scene.
I’m using the Law of Proximity (things that are close together appear to be more related than things that are spaced farther apart.) for many elements in my game. My Oculus Go won’t be able to handle millions of grass leaves, so instead I make a couple of grass patches, by bundling a dozen of individual leaves. Having them near each other, gives it the impression of it being grass.
For my snake, I have a couple of spheres and/or pyramids near each other. Here, the Law of Proximity in combination with the Law of Closure (the mind automatically fill in the gaps to perceive a complete image) are at play.
These are the shapes I’m going to use
Color palette extracted from Kuwahara autumn image
Since I’m more experienced with 2D animations than with 3D, I had to figure out how I efficiently can make my snakes move in the game. Doing this frame-by-frame by hand is very time consuming.. Luckily I found a way by using maths!
I made 2 splines (curved paths) using a sine formula. Then I used a cloner effect to clone 1 sphere and 1 triangle on their respective paths. Finally I could animate their begin and end positions so they would ‘move’ along their paths!
This is made in Cinema4D.
Sneak preview. Or Snake Preview?
Feedback for the future
During the meeting in which I’m supposed to present my prototype, I met Kai Giuseppin a lecturer at the HvA and Karin van Den Driesche, lecturer of this Visual Research course. They tested the prototype and this is the feedback I got out of it:
The introduction is kinda long and made him lose his focus. He told me that I could try and add some interesting elements to spice things up during the explanation, to get his attention again. This can be done with more (sudden) visual elements during the explanation.
The aesthetic style is cool, however the realistic image shown when the snake is unlocked make things feel “off” as this image has a huge contrast in terms of style. I can maybe try and play with things like the border or overall image style.
Needs more feedback on input and interactions. Maybe hold the snake in hands or put in basket? Simplest thing to at least implement is a sound to confirm when an action is made.
It wasn’t clear for Karin what to do to start the game. This might be a common problem for people who are not used to VR and/or video games. Need to make this more obvious.
As this prototype was made within a small timeframe many of the points which are brought up by participants were already known problems I had myself. However, it did give me some insights in what don’t necessarily needs fixing, but needs more attention. Like making sure that new VR users will understand the game. And to have a great flow in the game to hold the attention of the participants.