I’m very happy with the result of the project. As it works very well and looks good. If I had more time I would have liked to added an option for the user to upload a photo like you could on the teachable machine browser. And then even put the model onto a website for people to easily find. I really enjoyed making this project despite the hard part being done for me. I learnt how machine learning works better and through research and discussion about the project how heavily bias machine learning is and the potentially harmful impacts it could have.
Significance of colour
While I’m disappointed I couldn’t figure out how to add interaction to the experience I’m very pleased that I was able to make a 3d space that someone could explore as I have never made anything like that before. Also learning how to use unreal engine has been a great experience as I haven’t used software like it before. I would have liked to add more rooms of different colours as I learnt a lot about colour psychology. While I knew about it before this project I didn’t realise how scientific it was and how it could really impact a person. It made me look at real life situations like the government elections how the colour of the party might impact what you think about them. Especially in places like the US where there is a heavy emphasis or red and blue. And also the examples that Haller talked about in the book baker Miller pink and kinnick stadium.
Both projects have potential to turn into interesting projects if fully realised and I think I would like to take them to that point if I can.
I have encountered a problem my laptop has to go in for repair and I didn’t save my files onto a cloud. I don’t think I’ll lose the files but I am not sure when the repair will be done by. So, I’m going to try and finish both projects as best as I can but this has set me back a little bit as for example, I’m going to have to recollect the ‘is it cake or is it real’ data if the laptop does not get repaired in time for the deadline.
Tuesday
Is it cake or is it real?
I have managed to get access to my files for the ‘Is it cake or is it real?’ project. I’m using The Coding Train’s video on Teachable machine as reference to put my model into p5.js. I’ve started the p5.js sketch with ml5.js boilerplate version which has the latest ml5.js library uploaded already.
ml5.js boilerplate
p5.js code image classifier before improved user experience
P5.js code image classifier after improved user experience
Additions:
Loading animation
confidence score
User able to stop the classifier from classifying by pressing the mouse
I am happy with the code but I still have a day or so to add to it if I think of anything else.
Significance of colour
However I might have to do a prototype for the ‘Significance of colour project’ as I will most likely not have enough time to finish making a working interactive piece in time.
I have created the room in blender and I tried to use unity to make it interactive but I struggled to use it. If I get my laptop back in time I will import my 3d model into unreal engine and make the interactive part of this project.
3D Blender Model
Information to put in when making it interactive:
Turquoise can be uplifting and rejuvenating. Due to its vibrancy it can be stimulating for the mind.
Blue triggers mental responses and affects our intellect.
Lighter tones of blue are associated with mental calm, serenity and reflection.
Dark blue helps us focus and boost concentration.
Blue is the colour most likely to suppress appetite.
Light blue is soothing making it good for sleeping and dreaming.
Brands use it to promote honesty, integrity, approachability and expertise.
Thursday
Today I have managed to package my unreal engine prototype. I was unable to make it interactive as I was struggling to understand how to code in unreal engine.
Ramon Amaro: Machine Learning, Surveillance and the Politics of Visibility
Amaro talks about skin colour and facial recognition software. He uses the example of Joy Buolamwini and her work ‘The Aspire Mirror’ where she used multiple software libraries for the facial recognition. Buolamwini encountered a probelm during testing, the software couldn’t detect her face due to her skin tone. This led me to think about the issues with recognition software and bias. My ‘Is it cake or is it real?’ project uses Google’s teachable machine software to recognise the objects presented to the camera. However, I am in control of the data that the machine uses. So inevitably there is a bias as I pick what images the machine trains with.
If I had more time with this project I could have made it possible for other people to upload their own images to reduce bias as the machine would be learning from multiple sources rather than one.
In this week’s feedback Jazmin gave me a list of softwares I could use to create a 3d space for the ‘Significance of colour’ project as I had tried out a few and I was unable to do what I wanted. We also discussed how I could further push my ‘Is it cake or is it real?’ project.
List of softwares for the ‘Significance of colour’:
Aframe
Unity
Unreal engine
‘Is it cake or is it real?’ feedback:
Teachable machine does a heavy amount of the coding for you so to push the project I am going to refine the user experience of the piece. For example, having an introduction page/pop-up when the user opens the project.
‘Significance of colour’ dataset. This is a dataset of all the objects in my room that are blue.
Top right: A key as to what the ratings mean
Top left: What the objects were, Where they were placed, duration of placement, frequency of use, shade of blue
Bottom right: How many objects of each colour there was
Bottom left: Pie chart of the data from bottom right
For this project I also tried a 3d software called frame vr. However I found it difficult to use as it lagged a lot and I was unable to make changes to the objects like colour which is important for this project.
However, I have thought of other solutions. Such as blender which I have used before so it would not be too difficult to make a 3d model of the room I want to design based on my research so far. I have also found Gamefroot which another student had used for a task at the beginning of the year. With this I could make a top down rpg experience. However this would take away the immersive aspect. But I think that has a lower priority than the information the final piece needs to present.
For the ‘Is it cake or is it real’ I have collected all the image data I will be inputting into a word document.
how can human-machine interactions be thought about differently?what is the role of language/data/bots in facilitating or disrupting such interactions?how differently would you imagine things to be in 25 years? 50 years? And 100 years?
In the media human and machine interaction can be described as one being inferior and one being superior. Either machine being superior and human being inferior or vice versa. And that with machine being thought of being superior people fear a machine takeover. However like all things machines are not capable of everything and can make certain tasks less efficient. For example in today’s seminar the projector screen wouldn’t come down because there was an unknown problem. If the screen was mounted on a wall and didn’t have to roll down we wouldn’t have had this issue. But the projector screen is still useful to display videos, powerpoints etc. features projectors(machine) have over blackboards/whiteboards(human) So if the human machine interaction was viewed as machines can enhance human tasks innovations can be created without fear of a machine takeover.
Imagining what the future will look like depends on which routes humans will take. If we will continue to make our lives easier by enhancing all aspects of our lives to do as little work as possible or if we will try to use machines to solve important problems like poverty, lack of housing, etc.
Today we got feedback about our projects. Below is the questions and problems brought up and my answer and thoughts about them.
‘Significance of colour’
how will I display data?/what do I want the final piece to be?
I would like to create a more traditional piece like a painting as my other project is code heavy.
Project size
I originally wanted to record data on many colours. However, with the time frame this may be difficult to do and take up an unnecessary amount of time so I am going to reduce it to one colour.
‘Is it cake or is it real?’=
Jazmin showed me a software called Teachable Machine. Where I can input images into it and it will train image classification machine learning for me. So I will try making a prototype.
prototype
prototype
From this prototype I have learnt you don’t need a large data set to train the model. However, the more data you input the better your model will classify something. Also, I am able to copy the p5.js code of this model to alter as I please.
Use one of the methods from the lecture and apply it to one of your projects:
connecting and intervening in an existing network;
Put projects on the internet free to view
For the ‘Is it cake or is it real?’ project by putting it online for people to use it could help create a network of realistic cake artists from different parts of the world.
For the colour project by putting it online it could help inform a widespread of people about colour psychology.
For my colour project something like mez breeze design VR gallery would be a good source to draw inspiration on. As the final result for this project would be informing the audience about colour theory and an interesting way to do this would be to create a virtual space for the audience to explore as colour is best shown visually. Also a virtual space can be easily shared on the internet. however this is dependent on the audience’s access to a laptop and how easily that device could load a virtual space.