After Effects Multi Screen: Research, Screen 1 & Outcome

When approaching the second screen, I wanted to create something that resembled Artificial Intelligence which acted much like Iron Man's AI, Jarvis. I wanted it to reflect on Screen 2 by saying things like ''REBOOTING SYSTEMS''. But it had to fit with the actions of screen 2. So I looked back on the Screen 2 video and came up 5 phrases that would work:

1) "REBOOTING SYSTEMS" : The Matrix coding would then start raining down the screen to show that the system has now been started.

2) "ACCESSING USER DATABASE": The Figure would then appear from the centre of the circle, originating from a line of code. This would show that that specific line of code would make up that person.

3) "IDENTIFYING USER CREDENTIALS": The coding would start raining down over the rings and the figure to highlight that it is all the data about that one user (the figure)

4) "ERROR ACCESSING USER DATA": We would then see a mass overload of code. This represents the crazy amount of information that is online about this one person

5) "ACCESS DENIED":  The person then turns back into code and disappears amongst the rest of coding. 

Before trying to attempt in creating an AI effect, I wanted to look at different ways films have depicted AI.

Most films I came across depicted artificial intelligence as physical robots which is something that I wasn't doing. I wanted mine to appear as if it was behind a screen. For this I remembered a few films that had done this. A main film I looked at was Avengers Age of Ultron (2015). When looking at this film I really looked at the way Jarvis and Ultorns systems were portrayed and depicted. They emphasised that the physical form of the AI did not matter as the information was all online and that is where the power was held.

I really wanted to ensure that my AI effect didn't not resemble a robot. but was more pixelated and behind the screen To do so I found a tutorial of how I could make an AI type effect over a video of a person who would be saying the text above which looked like close to what I had imagined:

Through following this process and editing it to how I wanted I was able to create an Artificial intelligence effect. Through this process, I learned how to use the 'level' Effects and adjust the lighting. I also learned how to adjust and edit effects 

Skills learned through this process:

• using effects
• editing effects
• using a 3D camera
• creating a solid layer
• creating a point light on screen
• Using the pen tool to create a vignette effect
• the purpose of masks

However, it took multiple goes of creating this video effect as I came across a few problems. tThe first video I shot, the lighting wasn't right so the effect looked very odd over the video. As shown below, you could see he lighting was coming from the left hand side. This casted shadows across the right hand side of his face and across the wall. This made him look as if he was on a slight angle when in front the camera. The lighting for this first shoot was just the natural lighting from a window.

When shooting the actual video of the person, I made sure that I wanted no emotion or expression to be shown in the video as AI don't have the capability to have emotions. Furthermore, the only camera I had was an iPhone and I used a laptop screen for harsh clinical lighting on the face. 

Through this I was able to turn a video that looked like this:

 to this:

Screen Shots of work in progress

I made sure the main colour was the same of the Matrix code in screen 2.  I really liked the outcome. of this as it looked as if an. AI took the form of a human. I also really liked the Radio wave effect as it seems like it is emanating from a source. Furthermore, it is a reminder that the person in the video actually isn't real.

 I was very pleased with the outcome of the  video as I felt that it really captured an AI effect. Furthermore, I left the last 20 or so seconds as just the radio wave without the green face. I purposely did this as I thought that when putting the screens together I could have the radio waves in between each appearance of the face to highlight that though you can not see the face it is always there. This reminded me of the Amazon Alexa. Though it is quiet there is the idea that the little round tube is always listening.

This process of putting both screens together and editing on after effects was very difficult. Though I know I could have edited the footage on premier pro, I found that once exported, the Effects that were created looked slightly dimmed and not as effective. This is why I edited them together on after effects.

One big problem I came across was that it would all lag I wasn't entirely sure how to solve this problem. So instead I found that I didn't need to rely on it streaming to edit. Instead I looked at the Audio waves to see where he was speaking. By doing this I was able to see where I should cut the clip as there was a pause between each line that was spoken. I duplicated the 3D face layer and cut the footage to fit one different line of audio each. This meant I could place them on the timeline where they needed to be. I based this on the other screen with reference to the lines above and the positioning I said they will be in. I duplicated the same layer again, but this time cut it so it just showed the radio waves. I duplicated this many times and lined them up so it would loop and the waves matched. This ensured that when the face wasn't on the screen, the waves would be there. I also played around with opacity of the face layer to make it fade in and out each time.

I was very pleased with the outcome considering I learnt how to use After effects within the space of a week. Furthermore, I realised that although I started with a clear initial idea, that was quickly scrapped and instead, my idea developed along the way of making each piece. This was primarily because as I was creating I was learning new skills. So my ideas went alongside with what skills i had learned and could learn. If I had more time, I would like to play around with the audio. I would like to have a go at warping or distorting the voice on Adobe Audition as I wanted it to sound a little less human and more robotic. I did fin putting the screens together difficult but managed it in the end