• Location: Australia
  • Joined: May 14, 2014

Topic: A brave thought. Rotoscoping a sword fight

I want to create a sword fight that is as real as possible in terms of speed and timing.  The sword fight will be a key part of my next project and I don't want it to be plagued by swings that take too long or badly executed ease-in/ease out.  The best way of getting it right is through a technique called rotoscoping.

http://en.wikipedia.org/wiki/Rotoscoping

You essentially have live motion of the same action and then you copy/trace/duplicate frame by frame so your animation is as lifelike as possible.  With my version of Stop Motion Pro, I can do rotoscoping directly inside the animation program and can even put the live motion in the background of the shot like an onion skin.  All I need to do then is line up the body parts, take the frame, line up the next body parts, take the frame.  Its an industry proven but advanced technique.

The hardest part is finding live actors to film the live footage and to do the footage with the elbow fixed, no neck or knees and with shoulders that only bend up and down!!!

In the never ending quest to expand the brick film envelope, I'm asking if anyone has tried to do this in a brick film?  What worked?  What didn't?  The best part for me is with the power of the internet, it's not something I have to organise myself.  I just need to find two dudes or dudettes that like swinging broom handles around and think they can invent some choreography that will translate to a brick film.  Fancy stunts like backflips and what-not can't be rotoscoped (unless Jackie Chan joins the project!) but having basic moves on rotoscope creates a good reference point for speed and timing on other maneuvers.

Discuss.

Aka Fox
Youtube: My channel   Twitter: @animationantics
Best brick films: My selection
  • Location: United States
  • Joined: June 21, 2011

Re: A brave thought. Rotoscoping a sword fight

Umm, I'd say studying good brickfilming fights may be a better use of time and resources.
(And looking at good movie fight scenes wouldn't hurt either. You can get a sense of timing, cinematography and movements that work/don't work. But brickfilms are the perfect guide.)

Looking at good real-life sword fights may help a lot, but I highly doubt you could rotoscope something with much degree of success. The proportions are different, joints and flexibility are greatly reduced, and what may work in live-action simply may not in stop-motion. A less precise replication is certainly possible as evidenced by the multiplicity of scene and trailer recreations. But getting real folks to fight with Lego-like joints will be quite interesting and difficult.

You need to look at a frame-by-frame of some of the best Lego fights, and then practice and experiment a ton. Getting the easing in/out down right will be the biggest issue. I'm not saying you shouldn't recreate movie fight scenes as practice, but comparing brickfilm to brickfilm would be your best and most direct option at this point.

Youtube | Facebook | Twitter

Pritchard Studios's Website

  • Location: East Yorkshire, UK.
  • Joined: December 27, 2008

Re: A brave thought. Rotoscoping a sword fight

This is rotomation. Matching up the animation using reference, usually used by 3D artists to match up 3d elements in a live action scene.

It could work pretty well yes, takes all the fun out of animation, but it works.

Straight to the Biscuits

Rsteenoven's Website

  • Location: Welcome to Darkmoor
  • Joined: March 27, 2011

Re: A brave thought. Rotoscoping a sword fight

I don't think that it can work in the case of brickfilming.  Because minifigures have less joints, they cannot follow realistic movements.
Fighting is something you just have to try and hope it feels right.  I've done a few fights, but even I'm not sure they were perfect.  Sword and fist fights can be especially hard.

-Squid | Twitter | Welcome to Darkmoor

Squid's Website

  • Joined: August 14, 2010

Re: A brave thought. Rotoscoping a sword fight

I have used such a Rotoscoping process in the past with certain animations (specifically, it's how I learned my walk cycle back in 2008, or 2009) and have been very successful. For Imortalized, my current brickfilming project, I am planning on rotoscoping a lot of it - to better understand "human" movements, while also learning how to recreate those movements with minifigures.

Pritchard Studios wrote:

...The proportions are different, joints and flexibility are greatly reduced, and what may work in live-action simply may not in stop-motion...

This issue could be resolved pretty easily, in my opinion; What I plan to do for my brickfilm is to film the fight sequence once, trying to replicate each shot and movement as close as I can to the original, all done at 30fps. Then, I will break down that animation, converting it back to my standard 15fps, and then, I will re-shoot the entire sequence, with my own camera angles and shots, just with the same movements that I've copied. It'd take a little bit of your time, but, if you're striving for perfection, I'd recommend it - It's what I'm going to be using in my upcoming brickfilm, and, I'm not changing my methods unless I am unsuccessful!

Pritchard Studios wrote:

I'd say studying good brickfilming fights may be a better use of time and resources

I've never seen a good fight in a brickfilm that I've 100% liked. Even the fights in "Grace" look a little too cheesy to me - Sure, the animation is smooth, and, the characters animations seem to match that of the animation of the rest of the brickfilm. But, even at that, I hate it, especially the opening fight scene - Their movements and punches just seem so fake, and that really takes me out of the story - something that all art must avoid, otherwise, you loose your audience. (Don't get me wrong, "Grace" is still one of my favorite brickfilms of all time, and, It has a pretty descent ending fight, and, of course, one of the best dramatic storylines set to a brickfilm!)

But, to each his own - You should go with whatever you think is best for whatever you're trying to do. However, no matter what you do, I'd suggest filming some tests first, before you commit to one method of animation in your projects. Then, and only then, can you be sure that you want to rotoscope.

https://i.imgur.com/Z8VtGae.png
  • Location: England (sorry)
  • Joined: January 1, 2009

Re: A brave thought. Rotoscoping a sword fight

There is no realism in art.

This being said...it's worth a try. It might not work giving the fact that minifigures move differently to how we move, but I'm actually quite interested to see the results.

YouTube
Max, She/They

Max Butcher's Website

  • Location: california
  • Joined: December 27, 2008

Re: A brave thought. Rotoscoping a sword fight

it could probably work if you were to try to match the human movements as closely as you can with the minifig movements, even though it'd be difficult because as the previously mentioned problem of the minifig doesnt have any joints

this method definitely works better with traditional animation and cgi, also probably stop motion with puppets with joints similar to those of humans

i think a better way of going about it is just watching the scenes you'd rotoscope, study the way human movement works, and try to replicate it the best you can in minifig form

what could have been: jeffrey and the old man make some robots
                      art page -- tumblr --youtube
              bricksinmotion's #13th best curmudgeon