(04/03/2023 - 09/03/2023) - ROTOSCOPING EXPERIMENT

The first experiment that I wanted to play around with was 'Rotoscoping'. The original idea was for me to be able to rotoscope all of the movements that I had created references for. However, I realised that would take too much time and that I have limited time for this project. So I decided to set my goal to 1-2 movements for rotoscoping.

The reason for this was so I could have an idea of the benefits and the drawbacks that I noticed in this experiment.

ROTOSCOPING - ARMATURE ATTEMPT 1

For the first attempt, I decided to book the 'Bipid Armature' from the University. There were two to pick from, however, I chose this armature specifically since I liked the idea of the armature having limbs. The only issue that I seemed to notice was the lack of head. A solution to this, I decided to create a head by using plasticine.

Armature Bipid that I booked from the University

Once I started to attach the armature to the rig, I noticed that it wasn't the sturdiest of armatures. The part of the armature that attaches itself onto the rig was wobbly, so it kept on tipping over after several movements.

Due to this, I suffered a lot in terms of progress, since I wanted to finish the idea of rotoscoping in this week. I still feel as though I will be able to finish it this week, but not to the extent that I could've.

A screenshot of my workspace in Dragon Frame using my footage as reference.

Here, is my workspace in Dragon Frame, using my footage as a reference layer. I was able to change the size of the footage to suit the anatomy of the armature. I also noticed that the armatures don't really look similar to humans anatomically. So that was one thing that I noticed within the experiment.

Here was the 1 second animation I was able to produce with the armature. Due to it being extremely difficult to work with, it was extremely wobbly a jittery. Because of the armature being wobbly, I was unable to film more than 1 second (24fps) since it was unable to be sturdy from the waist. I tried to spend a few days to try to improve the condition of the armature, however, I was unable to do so.

So, I had to find a solution as soon as possible.

The only solution I could think of is book the other armature type from the university or try to book another armature of the same type. I had booked the armature of the same type, however, there were still issues with that armature so I decide to go for the other type.

ROTOSCOPING - ARMATURE ATTEMPT 2

The Metal Armature Kit that I booked from the University.

For the second attempt on the rotoscoping experiment, I decided to use the 'Metal Armature Kit'. The first thing I noticed after booking it, was to check the bolt on the waist of the armature. Thankfully, it seemed extremely sturdy so that was an upgrade from the previous. The drawback to this armature was that it was lacking the limbs that the previous had, making it feel less human. For now, I will keep it as it is for my experiment, however, once I finish with my experimentation of rotoscoping, frame rates and magnets and rigs I will decide to add some layers on the limbs so it seems more human-like. To add some character to it, even after adding a head.

ATTEMPT 1:

Here was a little experiment that I did with this armature for the rotoscoping experiment. I wanted to try something simple first before going on more complicated movement.

TIMELAPSE - ATTEMPT 1

Here is a screen recording of the first attempt that I had done. I noticed while doing this, that it was extremely confusing for me to focus on the reference footage as a separate layer on top of the armature layer in dragon frame. Because of this, I decided for my second attempt to try using rotoscoping but having the footage as a pop up video.

What are things that I like about it?

I liked the movement of the right leg moving forward as well as the movement of the right arm swaying upwards and back down.

What I don't like about it?

I felt as though I should've tried to add more weight on the left arm, since I wanted to create the idea of the staff hitting the ground. So, next time I should experiment more on weight and emphasize the idea of the arm holding the staff.

ATTEMPT 2:

This was the second attempt that I had done for my rotoscoping experiment and I believe there was a lot of things better with the second attempt.

What are the things I like about it?

I really liked the movement of the leg moving forward in this version since it was a lot smoother.

Not only this but I preferred the right arm movement as well due to the same reasoning.

Another thing that I noticed was that I started to focus more on the armature's silhouette throughout the movement instead of copying frame by frame through the rotoscoping technique.

Lastly, I really liked how the left arm had more force when it came to the movement of mimicking a staff hitting the ground. To make this more effect for my final version, would be to use a prop similar to a stick to make it feel more authentic and realistic.

What could be improved?

I believe the upper body could have a bit more movement to it and be more expressive throughout the movement.

Another thing to note is that for the final iterations, I would need to create an extra layer on the armature to mimic muscles on the body parts.

CONCLUSION?

What I have noticed while doing this experiment was that I ended up preferring using my video footage as a reference rather than trying to copy it frame by frame. Since I tried to copy it frame -by -frame in my first attempt, it felt lifeless, stiff and as a general rule felt extremely time consuming since it was filmed on ones.

Additionally, I was also unable to keep the reference footage as a lower opacity layer when trying to animate since it was extremely confusing for me to keep track on the armature as well as the footage, which effected my experiment.

Using my reference as a pop up video in Dragon Frame helped a lot since I was able to see the body more clearly and was able to experiment more with the silhouette of the body. I even tried changing the posing a bit since I didn't like how the reference footage had certain poses.

I noticed for the rotoscoping experiment, that it would be extremely useful to rotoscope the main key frames and then just work based on the reference.

It was really interesting for me to see that I will constantly keep changing things and that not everything is going to be finalized the way that I wanted them to be.