Post Production Techniques. Irony of analyst. Deduction

Final Piece:

---

Finding an idea

To be honest, I haven’t even started to work on my post production since last month. I was too busy with my animation project. But also I wasn’t sure what I could do. The brief was pretty much the same as the computer graphics unit, so it meant that we could do anything that would include green screen, rotoscoping, after effects, motion tracking, 3D modeling, colour correction, sound design and many more.

I was hanging on facebook and talking to Pablo, when i found out he wasn’t sure about what is he doing for it as well. So we decided to work together, but we didn’t have any ideas.

2 weeks passed, I was working on my side project for my friend Rohan (his photoshoot) and also was working on my animation project.

I was searching for inspiration.

Usually the inspiration is coming from a music for me. I always find something I want to film about through my dreams, or when I’m chilling and listening lounge or downtempo music. Also I really find that selecting a music for the video you are going to make is one of the most important things. I’m not sure how professional artists work, but for me, it works perfectly. I develop the idea, i search for music, i edit. Sometimes it works other way: I FIND the music I’m inspired by, I develop the idea, I edit. The editing should be fitting to the music(or opposite) because that is what on 50% motivates you to watch the video. The commercials, films, video games, everything is connected to it. Where “Once upon a time in America” would be without it’s music by Ennio Morricone? I don’t think it would be a great classic film as it was then, and as it is now. If you want an example of recent films, i could say that “Inception” has a pretty decent soundtrack which made an impact and ispired a lot of people. If you want a game soundtrack example, I would say Assassin’s Creed II made a big impression on me, just second part of the franchise, also the first one… but not the newest ones. I could also name “Machinarium” and “Journey” or even “Uncharted” series. But mostly I like the old games soundtrack like “Neverhood” or “Worms Armageddon”, “Adventures of Lomax” or “Donkey Kong Country”. That’s what makes us to remember games and films even better.

-

By the way if you are a fan of Donkey Kong games you should check out “Kong in Concert”. It’s a Donkey Kong Country arrangement collaboration. Orchestral music with is brilliant.

-

Anyway, recently I was watching a show called “Sherlock”. I was really impressed by this version of Sherlock. I’ve seen pretty much every visual version of Sherlock and this was one of my favorite, despite how “crazy” and “drug addict”-like he was in it. But he wasn’t a drug addict, of course no. He was really bored and couldn’t wait for a new case to solve. Maybe it’s a little bit weird, but at least its really funny to watch… so, ehm… what was I talking about? Ah yes, the inspiration.

These videos may explain you how Sherlock’s mind works, and his whole theory of “Deduction”, and you may see something very simliar to our video:

---

But before that i was thinking about how would virtual environment work. I was developing the idea about a guy who walks around the town and scans people in even more detailed method. Their whole lifetime was in his eyes, all their bad or good habits, what they like or dislike etc. Starting point of this idea was this video by Nickelback.

We could track people and put some information above them instead of numbers…

I told pablo I wanted this to be like a “Game-like environment” but in a real life, maybe even to make a guy to be a robot. Also we don’t need this idea with “information above heads” for the whole video and we don’t need a strong narrative for it, we just need to show what we can do. So we can show different techniques in the video.

I told him that we could do a dialogue as well but with adding some humour to it. I wanted to refer to “Heavy Rain” or “L.A. Noire” games dialogue system. The question is “what if that was in real life”? For example when you talk to people in the game and you have a chance to choose an answer, you can have as much time as you want to choose it, so you can think about different possibilities of your answer. But if that was in real life people would think that this guy is an idiot, so we wanted to make a funny dialog of main character and a girl, like if he is trying to hit on her. BUT the funny part is he is thinking too much about his answer, or even his answers don’t make sence because he has too much answers and questions he could say so he chooses the wrong ones.

Pablo wanted to focus more on statistics and information we could use in the video. After a while the gaming idea was declined. We made it a lot simplier task for us to make a main character a man and use deduction and possibility method. We thought that main characters opinion could have been even more funny. He could not only say actual facts about person he’s scanning, but his opinion as well. But anyway Pablo spent some time on finding statistics about people we could use in the video.

My idea was also the beginning of the video, Pablo had a different thing on his head, I honestly don’t remember what it was but in the end we liked the “changing background” idea and we used it. Also the first idea was filming Merlin just normally without any reverse. But I came up with a “reverse” idea. I thought that would look much more convincing when his beginning of the day skips even faster with things jumping into his hands instead of him grabbing them.(fork, knife, notebook, pen, bowl, jacket)

We didn’t know what could the video end with. In the end we wanted our character to scan the beach and people on them. If people would play beach ball we could do some more things like ball trajectory and things like that. But in the end, because of the weather we needed to shorten this scene and do something different, i will tell about that a little later.

We had the idea ready, we had most of the things planned, but still we were unsure about the outcome.

---

Filming

Pablo asked Merlin to act in our video. We didn’t need a professional actor for this scene as me and Bertie needed for “Escape” video. We had my Sony HDR-CX700VE and also a Canon 5D mark II for one day. We went to town center 4 times in total to shoot everything. Mostly it was because of the weather.

First time we shot The park part with him scanning a baloon, bin and tree and also a third-person part with him getting up from the bench and starting to walk. We shot some of the beach shots of him looking into the sea but because of the weather we couldn’t do much because it was cloudy, when we needed a sunny weather.

Second time we refilmed the whole beginning of him looking at the baloon because it was more sunny, the way we needed weather to be, and filmed the whole park long shot when he scans people in it. Also we filmed the golf shot with some random people. We just timed it right. We were lucky enough to have a lot of people in the park because of good weather and because of some event! We decided to film on my camera because it had Optical Steady-shot system in it and it didn’t shake that much as it would be shaking with a DSLR. It was me who did the whole long shot of it. Also I filmed the first person beach shot of him looking at carousel, jumping kids and ice-cream kiosk. Pablo did nearly most of the DSLR shots. But we haven’t filmed the ending because of the weather. We filmed something though which we didn’t even use.

-

We had a tough time with a DSLR by the way. We both were unexperienced with Canon 5D. I think we should have some tutorials on it or on any DSLR in second year. I hope Elliot will help us with this.

-

Third time I met Merlin and filmed the whole ending in 15 minutes. The whole thing starting from him looking at ice cream.

---

We had all the shots except the beginning (with Green screen). We filmed it in the university, in our class with booths. I think Phil saw the whole process and even asked us: “Why are you throwing cutlery around?”. It took Merlin a while to master this whole reverse shot.

---

Editing

-

Let me talk about editing itself.

I use After effects only for defined shots to add visual effects and do things that you can’t do in editing programs like Premier Pro, Sony Vegas, Avid, Final cut etc.

I started to edit different shots as soon as we finished filming.

When last time I mentioned that I use different software for editing (It was sound and video techniques post) Phil told me that I should use another software which is “Premier Pro” because it’s claimed as a professional-industry software. I’m not sure if Phil deducted a few points out of my mark in the end and I’m not criticizing anything whatsoever. As a matter of fact I aprecciate his advice and I took it into consideration. BUT, I would like to tell you why I decided not to use it.

I don’t use Premier Pro because of it’s “After effects”-like interface. After effects has a really uncomfortable interface in my opinion, but it’s a really powerful program and I’m forced to used it because it’s the only one I actually know how to use. I’m thinking to switch to “Nuke” anyway.

I’m not going to talk about everything I hate in After effects, but the only thing that i’m going to mension is that Premier Pro has pretty much the same interface as in after effects which is not excusable for me because the actual edit is supposed to be 10 times faster than in visual effects program(after effects) which doesn’t happen in Premier Pro, but it happens in Sony Vegas for me. That’s why I’m using Sony vegas.

I just want to say that it’s not about what you use, it’s about how you do it. But if you want some statistics, try searching on youtube “Sony Vegas” and you get 226000 results, when if you try to search “Premier Pro” tutorial you get only around 34000. You may ask “why?” and I would say because Sony Vegas Pro has comfortable interface but at the same time, it’s really powerful, you get more options and manual controls on render settings. Also amount of tutorials is enormous, it’s just simplier to use and easy to start with.

Maybe if I will be working in video production company where everyone uses Premiers and we will need to pass around projects, in that case, i would learn Premier. But for my projects I would use Sony Vegas Pro.

I think you would understand me, because I’m not the only one who uses different software not the one we should use on the course. Many people did they animation projects on 3DS max instead of Maya, and sometimes it looks even better than on the software we supposed to use but it doesn’t mean it will look the best because It depends on a persons knowledge of a particular program. And I find it wrong when people say that I shouldn’t use it because it’s not professional. In my case I don’t find Premier Pro as a “Pro” and the editing process is really slow.

I’m not telling you that no one should use Premier Pro, I’m telling that people should use what they are good at. If they feel that Premier Pro is the best for them, I think they should stick with it.

One day I would switch the software just in case it would be necessary. But if you ask me which software will it be, I would say – it will be not Premier Pro.

-

---

The filming part didn’t take that long. The hardest part was post-production itself =) And, just to make you aware of this, the sound effects research and editing took us more time that the actual visual editing, as you can see on screenshot above.
The visual effects with text editing was made by us both:

Intro: Me
Baloon shot: Pablo
Scanning People and Golf: Me
Spinning Carousel: Pablo
Boat shot: Pablo
Ice Cream: Me
Watch: Me

The first thing i did is made the actual intro, which on technical side looks like this:

Here are some of the screeshots:

Here are some more screenshots:

The hardest part and the most time consuming was tracking people. The tracking on X and Y axis visually doesn’t provide the best results because when the people go out of frame on the left or on the right, the text goes “to the left” and “to the right” instead of scaling and going into the camera. We even tried to track scale, but it doesn’t track well. So in this case i found a solution of that problem.
What I did is I tracked X and Y axis but made it work only Y axis only by adding an expression [250, this_comp.layer("Null 1").position[1]] to the position and I animate X and Z axis by myself. That’s where i did lots of keyframing. Took a while to make everything look good, but still worth it.

Pablo had an experience about type font since he made his computer graphics unit video and he suggested to use particular font for our video. I spent a huge amount of time finding sound effects that we could use for our video and we spent a lot of time editing them. We were also thinking about facts which we could use for people in our video which could be not too offensive, i think we came up with not the best ones we could ever think of (We were really tired by the end of the week) but I think everything turned out pretty good.

As for the music, I think everyone has noticed that we used one of the songs by LCD soundsystem which was used in GTA 4 UK commercial:

I think that was the only song we could use which sets a tempo and a special “badass” mood for scanning people

---

Evaluation

In my opinion we did a really good job on this video. I’ve never done that much work on text, type and transitions of it (when text jumps out or disappears)

I can’t even say what could we have done better for it. But if i’ll try to think about it, I would refilm the whole thing with a better camera. I still don’t know if there is a way to prevent 5D from shaking that much, or is there a build-in steadyshot technology on any DSLR or lens, the shoulder rig we had didn’t help that much.

Pablo did a very good job, I enjoyed working with him, he has a lot of experience on different things in editing and most of the times I needed his opinion on everything. His editing is really good as well. We taught each other very useful things because we had a lot of our own strenghts and weaknesses.

I’m so happy again that our teachers liked our work, can’t even describe how. Well, I’m glad that you like our work!

Here is the behind the scenes bonus video :3 :

Ged.

Principles of animation. Evolution of Dance

Finally, after all tests, and ideas that seemed impossible, we’ve got it…

This Project was made by Bertie Sampson and Gediminas Jurga(me) as a group project.

---

Preproduction

When we had a brief, i really liked how Liam described what was the idea of this unit and what does he expect from us. The basic idea of it was to develop a character. The video in which a character expresses some emotion through movement and gesture.

The whole day we spent watching examples of character development, basic movement. We examined characters nature and temper and what it could have been doing in the scene, for example fear, shock, affection (as its said on the blog.)

I was quite confused on what could i do: whether learn some maya and do the piece on it or just stick to the good old stop-motion.
At that time, i was inspired by this video, which represents that author used green screen and post production. But the whole idea of the video is really good, and the way toys were used is brilliant!

I was quite happy with this idea for a while. I wanted to make something like a stop motion with toys and, maybe, clay. But then something has changed my idea. I honestly don’t remember what… Maybe i was surfing on Youtube in search of inspiration and watched a lot of videos about face tracking and motion-capture.

Examples:

Then i found videos like this:

I was quite interested in the the whole idea of cheap mocap. Everybody even has this technology at home which can provide stuff that no one can actually think of… like Microsoft Kinect and Playtation Eye cameras. The mocap becomes so popular nowadays because this technology is used in video game development and film development (If someone has seen James Camerons Avatar)

Some of these example videos show that some of the capture was done with a cheap method at home with a simple web cameras and stuff like that.

I was fascinated because if i would be able to test this technology and do a final piece with it, that would be awesome. But at the same time i was quite nervous about it and unsure if I can do it without any help and equipment.
I’ve decided to do some of the tests on this to find out if its easy enough to pull it to the crit.

---

Preparing ourselves. Ideas.

As i said before i wasn’t sure if i can do it without any help. My idea wasn’t strong enough, i didn’t have any idea of what am I doing and what kind of narrative i could build with this kind of technology, because it was extremely close to acting itself.

I set myself a goal to make a body motion capture first then make a model, apply the mocap data to a model and then, if i will have enough time, maybe I could do a facial capture.

I was quite happy with final piece for Sound and Video production, which i’ve made with Bertie. He is a nice and creative person with whom i really enjoyed to work. So I asked him if he’s interested to help me out or even make the project together. It was a day when we had tutorials with Elliot. So it was the right time to start working on it.

Bertie was very interested in my idea and we started developing the idea.

On the technical side, we tried to count the time of the rendering our piece could take. Yeah we were curious if we could render it with ambient occlusion, lighting, basic models and the best settings which maya can provide these days… We thought we are going to make 30 second video, as it was said on the brief…

-

Tutorials that Elliot and Liam were providing us with were actually really helpful. Some of the basics they showed to us we used in our final piece, but still we needed some more deep tutorials on how the maya interface works. There were so many things we didn’t know how to use like the whole section of Human IK which was the basic thing we were messing around with and without such knowledge we could not have been pulled our project to the crit, and also the Outliner windows which is one of the important things you need to know.\

-

We were sucked deeply into this maya problem later on, but obviously, we couldn’t wait to do the tests on mocap. I asked people around if anyone could provide Kinect or PS3 eye cameras because i left my PS eye camera back home in Lithuania but, we definitely needed more than one. I’ve posted a request on our group facebook page and at the same time, with this post i’ve exposed what was our idea of the project… The thing is i wanted to keep our project in secret so we have a surprise for everyone in the end, but its too hard because we needed some assistance =)

A lot of people replied that they are happy to provide the cameras and Kinect sensors. Also some people said that they are doing the same kind of stuff (In the end I didn’t see a mocap video in class on the crit other than ours, maybe because it was really hard to make everything work in the end)

---

We sticked to PS eye cameras because it can capture 60 frames per second and it has wide angle lens when kinect does only 30fps and im not even sure about what lens it has. But if we decide to do something fast-paced using fast movements, kinect would not be able to handle it and the mocap is going to be impossible to capture because of low shutter speed.

Han Ying Looi helped me with two PS eye cameras, his and his friends, also Euan McQuisten from second year DMP course helped me with third one. I didn’t know where to get more than three… Usually more cameras means better – less jerky, jittery motion capture.

I’ve decided to use ipisoft for capturing because that was one of the most accessible 30 day trial free software which had a good strong reputation of “cheap mocap” on internet.

We were prepared to do a first test…

---

Work in progress

As i had not enough space in my room because its a flat, Bertie said we could do it in his living room.

The software we were going to use is ipisoft as i said before. It works only on Windows PC. As I am a persistent PC Windows user (Hate macs for some reason) apparently i was the only one who was able to use this software for our project. But the problem was that I have a desktop PC, not even a laptop. And none of Berties housemates has PC. So we had to bring my computer with all its parts to his house several times afterwards. (I think 4-5 times, to record our project and do tests.)

Another issue was the software itself. This software is really hard to make it work. Whole day we were messing around with its installation because it didn’t want to launch. We had so many problems with it in the beginning and to make it work we did a huge amount of research. We reinstalled drivers twice on my computer, reinstalled the software more than 20 times, read forums, uninstalling other applications which could conflict with this. We solved it eventually. It was an issue with administrator rights on windows and we needed to launch the file in the ipisoft system folder instead of an actual shortcut.

Finally we WERE ready for it and everything seem to be working so we set up our cameras on tripods, marked the space which is visible on every camera and where the character can move around (As you can see the white tape on the ground) and did the test which resulted in this little video which i posted months ago:

We were quite happy with it!

---

The idea and the music

It was time to come up with an idea. We literally didn’t know what to do because there were so may things we could actually capture with this technology!

One of the few draft ideas was to capture a tai chi. As long as its really slow process, the movements are really slow there will no problem capturing it easily. We wanted to make the whole room with robots who do the same thing like practicing their kata or a tai chi or stuff like that.

That was quite boring and we came up with dance evolution. Im not sure was there an inspiration all… and im sure that Bertie came up with it, maybe he had something on his mind, but i really liked it. We’ve watched some videos of how people actually understand the evolution and what does it mean to other people.

We did a huge research on a music. We were focusing on the iconic songs and what WE remember from our childhood in first place. We used songs that we were grown up with. The process was very interesting… probably one of the most interesting experiences: one day we were hanging at Bertie’s house and just thinking of what songs were popular through ages, playing it out loud on speakers so his housemates were interested in it as well and helped us to choose the songs. Those suggestions were really helpful in the end. We came up with the list of song we might use:

The Who – My Generation;
The Beatles – Help;
Jimie Hendrix;
Spice Girls – Wannabe;
MJ – Thriller;
Bob Marley – Jammin’;
Led Zeplin;
Pink Floyd – Money;
Darude – Sandstorm;
Limp Bizkit – Rollin;
Metallica – Enter Sandman;
Nirvana – Smells like team spirit;
Queen – Bohemian rhapsody;
Deep Purple – Smoke on the water;
Abba – Dancing queen;
Elvis – Hound dog;
New Order – Blue monday;
Kraftwerk – We are the Robots;
Vanilla Ice – Ice ice baby;
Daft punk – Harder faster better stronger;
Fatboy Slim – Weapon of choice;
ACDC – Thunderstruck;
Aqua – Barbie-girl;
Freestyler – Bomfunk MC’s;
Scooter – How much is the fish;
Vivaldi – Spring;
Haddaway- What is love;
Las Ketchup;
Ghostbusters – Theme song;
Scatman John – Scatman.

But it was impossible to use every song so we needed to filter this list and choose only some of them to fit the video into 2 minutes (at least) because final peace should be 30 seconds and maybe not that long as we wanted it to be.

So we started the edit. Bertie did the first edit of the songs it was quite cool but the order wasn’t quite coherent because these songs were composed in different years.

So I’ve tried to make my own edit of it according to actual chronology and also added some effects like reverb and vinil scratch:

Bertie suggested to use different parts of songs (For example in Spice girls song and Blue Monday he wanted to use not the chorus parts)

So I did the second edit of it

Then we decided to use another parts of songs again and also delete scatman song because it didn’t fit well. So there we have last edit and the final one which we used in the video:

We were quite happy with it!

---

Tracking (Really long process)

The tracking went wrong for the first time. The thing with ipisoft – it needs all the measurements like the distance between cameras, their height and the angle around the capture point.

We had all this, but one of the cameras – because of a stupid tripod – kept panning down at the end of the capture so the track went very frustrating and virtual cameras on ipisoft were messed up. I positioned cameras by myself and it took me whole day to understand how it works and make all the virtual measurements right.

It was tracking for 2 days basically, and i needed to watch this process. Literally sitting at home for 2 days straight. 60 frames per second. 8100 frames in total. 6 seconds render per frame.

You may ask me why was I watching the process? Its because at some points some part of the leg or arms went wrong and didn’t track well. If that happens in one frame, the next frame is going be the same and the whole tracking is messed up so if it went wrong at some point, i repositioned the virtual skeleton again by myself and put it on track again. Sometimes if you reposition it it goes wrong again because it doesn’t see the arm in the shot, or it blends in with the body color so i was forced to do that frame by frame. That’s why sometimes you see some jitter because it didn’t track well and in the virtual ipisoft space it was hard to understand what axis was wrong.

But we hid most of the errors with camera angles afterwards.
Here’s some of behind the scenes footage with ipisoft final tracked peace.

---

Developing a character, Maya (All the frustration)

We needed a solid choreography for this. So we asked some dancers from the uni to help us out. But because of that, the whole process was slowed down. We asked 2 people and no one was able to make it on defined dates. The most annoying thing is that they were telling us that they are not able to do that in last moment right before we were agreed to meet. So 1 month passed without any progression until me and Bertie were frustrated and decided to do this on our own.

We made Bertie’s housemate and another friend to do some dancing as they feel will fit to the music we edited. We were observing and writing down some of the dances they did. That was quite fun.

So basically we didn’t even had a script for every dance, each dance was quite random and unpredictable.

The last dance was made by Bertie’s housemate Roman. The capture was done on the next day.

---

The model that we were going to use was the one we got from CreativeCrash.com website. It was a freeware model and we were quite happy with it. But nothing was that easy. This model was rigged and didn’t have a skeleton in it. We had so many issues and didn’t know how to insert a skeleton in it. Also when we tried to delete the rig we couldn’t select the mesh again… so we just started to search for other ones.

Heres and example of another model we tried to use, but it wasn’t quite good for our project. We needed something that was looking like human. A robot basically.

---

Maya, maya, maya

We did a lot of work to make this model able to work with. We needed to reposition all small parts of arm mesh through outliner because when we delete the rig the model looses the T-pose and put its arms side by side. All the mesh parts were not connected so it was extremely time consuming to position everything correctly by working with numbers in attribute editor so everything looks mirrored and all the parts positioned correctly.

Later on I did some tests on referencing the Mo-cap data on other rigs and models with HumanIK in maya.

There was a small chance that Maya would not be able to work with those files and we could have been forced to use MotionBuilder or 3DS max. But it went out ok. Not perfect but still something! This model was generated in ‘Make human” software.

We figured out the whole Human IK thorugh this VERY USEFUL tutorial. I think we wouldn’t have made this project without this guy. Big thanks to him for his very helpful tutorial on mocap, human ik basics and skeleton characterization.

---

How the work was separated?

When we nearly finished the project we wanted to position fingers at some point as well, so that was my part of the maya work. But it had a problem. If we apply mocap data to a model – we couldn’t move fingers because they have their static position for each frame, so if we change one frame the next frame is going to be as it was according to mocap, so it wouldn’t change. We couldn’t apply it to all frames so we tried to disable them from mocap data. But if we disable fingers from mocap data the whole hierarchy of the fingers was broken – so if we tried to move one joint, the other joint which went after this one and was parented to the one before didn’t move for some reason… But eventually i found out the solution it i was forced to move each joint of the hand and keyframe each joint as well… Took a while, i must say.

I did different kinds of hand positions:

…and Elvis dance – pointing finger as you’ve seen before =)

Also I did the head animation to the rock song, but it wasn’t made in maya, i did it in ipisoft while it was rendering frame by frame because, i think it was much easier to do it in ipisoft.

As for Bertie, he was researching more on textures, lighting, blendshapes, reflections, render settings and all stuff like that. I didn’t know how to do all of this stuff so his knowledge was really helpful, I guess he has all this information on his blog, if you are interested.

As for the camera movements we were sitting for two days and animating three different cameras for 3000 frames. Yeah originally it was 8000. The framerate needed to suffer and we cut from 60fps to 24 fps so we don’t need to render that much.
Some of the cameras did a weird movements because we couldn’t get how the point of interest of maya cameras work. That’s why you saw this weird movement of camera rotating and going between robots legs ;D We couldn’t control this part and we meant a completely different movement, but i think it came out pretty well… We wanted to add to our video some kind of feel. Yeah we could just put cameras at different points and make them static, but i think it looks much better with panning and moving around cameras.

---

How the final piece was meant to look like?

The beginning part in the final piece was ment to be a little bit different. All the spotlights and dark room was our original idea, but the beginning part, when robot is looking around and he is a bit unaware whats happening, at the part when he moves his arms down slowly, there should be a disco-ball coming down and start spinning as the music starts, but we could make one and whats worse we couldnt make the reflections work on the ones we downloaded.

Also the floor we used was textured like white granite but we couldn’t make the reflections work right as well so we had THIS issue:

For some reason the model didn’t want to reflect in the floor as a whole thing. We were messing around in renderer, reflections, texture properties. Nothing. It reflected only parts of the arms only for unknown reason. We left it to be like that…

---

THE RENDER

We had the final piece ready. But we knew that render was going to take a huge amount of time with the good high quality render settings. And GOOD THING we wrote all the frames we used for different cameras. (That was a good use of iPad by the way)
(Picture on the right)

So Bertie found a tutorial on how you can batch render through a command line.
http://www.jawa9000.com/Technical/batch/batch-rendering.htm

I’ve tried it and after few experiments with a script which looked like this:

It worked for me on my Windows PC, but it didn’t work for Bertie on his Mac for some reason. There were some sort of permission issues and conflicts with paths which i’ve experienced as well but solved.

Basically the idea of command line rendering is that you don’t need to set your batch render every time if you want to render different scenes. You can set up a script, write every line for shots from frame to frame (as you see -s -e means start frame and ending frame) also set up a render setting (-mr means mental ray rendering) and the path where you want it to save. So when one line render has finished, it starts the next one. Having this script will give you an opportunity to leave it overnight instead of setting up each render.

So we just wrote the whole script with frames we needed and rendered it in a pretty fast way I set up my PC over 2 nights and we were done. Bertie has rendered some of the longest shots as well but he needed to set it up every time.

We were done here. After that we used after effects and basic editing.

---

Editing

I did the editing, added some ambient sounds, added the projectors sound when they are turning on in the beginning. Also i wanted to build a narrative for that and i was quite inspired by the GladOS voice from Portal 2 so i found a text to voice generator and added some effects to it like reverb, pitch and etc. I think it turned out pretty well.

---

Evaluation

I’m really happy with our peace. It was so exciting watching our course mates liking the work and some of them even said “This was the best one”. I liked nearly most of the works my course mates did! I was really happy to work with Bertie again. And again we produced really nice work as we did with sound and video production. I hope we are going to do more projects in the future because our final pieces of work turn out really good when we collaborate. We had a lot of issues with this, but we pulled it out to the crit. I really enjoyed with a process, and I was quite interested to learn maya and the whole HumanIK. I think i’m going to learn more of it in the future. But for this year, I think I’ve learned enough.

Nevertheless we had a lot of issues with it. Despite all the problems which we managed to solve, we couldn’t do a different setup of the cameras because we didn’t have proper USB extension cables. The ones we had didn’t work for some reason. With a different setup we could capture much more high quality mocap data without any jittering. So without any extension cables cameras were not so far away from my PC and they were not “around” the person.

The whole idea of my project was to set a plank and show people that we don’t need an expensive equipment to do something better. It’s about how we use stuff that we already have. We have so many hi-tech devices in our houses and we don’t even realise what are they capable of. It’s not about what you use. It’s about how you use it.

Thank you all for good words about our work! We worked very hard on it and we are really pleased to hear good things about it =) Liam asked us to make a class about it, I’m very happy to show the process we went through. We did a lot on this project, went through all the problems and we can teach someone in a simplified way.

As for the course, I would suggest to have PS eye cameras on our course to develop a unit which gives an opportunity to make a motion capture. That would be great! So people could book out cameras or kinect sensors from the uni and make their projects.

The post is final, but im gonna be editing some mistakes in it and maybe resort it a little bit later(tomorrow maybe). But for now, I’m off.

Ged.

Generalizing- Work Placement

It’s been a long time since I’ve started my course. Since then I’ve made quite a few videos(starting from proffesional work and ending with my own videos) and i’m really happy with the outcomes. Now it’s time for a work placement.
Phil provided us with different work placements and I think “Love love films” is what i was searching for. So i decided to generilize my work in this post.

Short films and viral videos:

Escape (This video is about what a man could possibly dream, living in industrial, digitalized and internet-based “stone jungle”. By the way it was my group project and I think we all did a very good job on this.):

60′s Couple around bournemouth photoshoot (Did this video for my friend from Photography course):

One more time (A video which is involving days of postproduction, and tells about a guy who is playing too much video games)

Jurassic Coast (It’s a video which was made in my own time while i was travelling with my coursemates):

Epic space flight (This video was made by me in uni while we were testing green screen. I think it could possibly be as one of those viral youtube videos):

---

Documentary: (This documentary I made when i was in high-school. I’m international, that’s why it’s in different language. But it has english subtitles)

---

If you’re interested you could check behind the scenes:

Escape – Behind the scenes:

One more time – Behind the scenes:

Also you could check my vlog. It’s something that I do rarely.

Motion capture test 2

Hellot there,

A lot of time passed since we’ve started the animation project and finally it’s coming to an end.
We have left to do some tweaks to the animation and the mocap data to make it look perfect. Here is one of the test that we applied to a custom model. It looks weird though so we are going to use another model instead.

By the way the second dance he’s doing looks wrong because his arm is too long. He should have held his leg instead.

60′s couple photoshoot – behind the scenes

I’ve been working a lot on the side projects this week. This video is “Behind the scenes” for the 60′s couple photoshoot for my friend from Commercial Photography course – Rohan Hande. I think it came out pretty well.

Cheers

Juggle video test (Not a uni project)

Me and Bertie were inspired by one video where the juggle ball was tracked and the path was applied to a camera.
We did a test of juggle and some nice levels correction which was done for tracking,(we wanted to remove it afterwards) but color correction remained because of it’s awesome look.

Here is it:

It’s not a uni project, we were just having fun. I think we are going to finish it soon.

Animation Project

Hello world,

Okay today we tested motion capture technique with Bertie for Animation project! It took us whole day to make everything work properly. But finally we made it and here’s a little test video we did.

This test took us a whole day from 12:00pm to 21:00pm. At least we know how to do it properly now and will do it better next time.

Still need to come up with a proper idea for our project. A little story or something that will create a narrative.

Escape – Behind the scenes

Hello there,

It’s been quite a while since we had our Sound and Video crit day.
Anyway, i’ve made a little “Behind the scenes” video about how we did our “Escape” video.
Check out if you interested!

GeDoZZZ

Progression.

Hello there,

After digital publishing hand in we had a lot of Maya lessons with Liam and Elliot. Those classes were really heplful, but I still need to figure out how to place joints and the whole bone structure correctly… I hope I will.

---

By the way, I came up with I very good idea about my animation project!
Basically I’m going to make an animation using mocap. But it’s not simple as it is. I researched a lot. I took me 4 hours to finally understand and figure out how to do that properly wothout having any mocap costumes and expensive Motion Capture cameras.

I’m going to use PlayStation Eye cameras to do that. It’s very complicated process and a lot harder to do than it seems. Youtube tutorials are actually not that helpful. I found a lot of forums, topics and guides of how to do amateur, cheap, but at the same time – high quality mocap using just PS eye camera or Kinect sensor.

I’m not going to describe the whole process here, but I will describe some processes afterwards.

---

Today we had lighting and green screen tutorial with Elliot. That was helpful even if I had experience with green screen before when we did “Escape” video with Bertie and Han.

We recorded some random things on green screen and needed to do some post production on it. Here’s a little video I did.

Gediminas.

Maya basics today!

Hello,

Today we had basics in Maya with Elliot. I’m quite happy that we learned basic things and tools to start modeling, experimenting for our new project. Here are some screenshots of the banana I made in it ;] This is really easy to use programm (at least for me, because I’ve never had experience with modeling before) Now I need to learn more about textures and frame-by-frame animations.

Ged