Personal Project. Researching Concept Art

Screen Shot 2017-05-17 at 12.08.16

As a disclaimer for the personal concept art project; during my process of studying the techniques of practicing concept artists and their work, I was attracted most to an early stage in production. This being when imagery is very conceptual and loose, my earlier works being exceptionally so, based around patterns and studies. My technical drawing of interior design and architecture has improved dramatically and I think all my studying of perspective has really benefited me. My body of work here is an attempt to create a mood, an atmosphere, an ambience. Therefore it is not nailed down finalised character design etc, though I appreciate that this another crucial aspect of concept art. It is rather an earliest stage in production, where characters are not characters and are more figurative. Seeing how a figure translates into an abstracted location.

It’s this area of design which attracts me to concept art.

Screen Shot 2017-05-17 at 12.08.16


What this personal project became; originally I had set myself a narrative. Though after attempt after attempt to stick to one setting I had decided for myself to draw in advance, I found that I had set myself up poorly in terms of artistic development. It became more of a focus to a general “dystopia” and general “figure” as representational elements, rather than a strictly laid out narrative to illustrate. Perhaps for me art and narrative should be made simultaneously, and each should feed off of the other. My final pieces are very abstracted, but I am personally satisfied with my movement from traditional to digital artwork. I find that my skills are transferable and that with the use of digital media I can achieve things in photoshop in regards to 2D image making that perhaps I couldn’t have done before.

To inspire and guide my development in this personal project; I chose to find inspiration from artworks produced within the concept art industry. I have selected a few particular artists whose work I find especially interesting. Looking for a common thread throughout; I’m not interested in concept art because it appeals to a certain genre per se. For instance some people may be drawn towards art such as fantasy or sci-fi. The common thread for me, I am drawn to artists who work is very technically impressive, I also like things that may be considered dark. The artwork I select for inspiration is based off realism, I’m not overly interested in cartoons. And I hope that there is a sense of surrealism; though due to the design aspect of the concept art industry, abstraction and surrealism tend to be experimental choices. Therefore they’re not always seen in commercial art.

I hope artwork such as this has inspired and influenced my artistic tastes.


Surreal “Sphinx” sculpture from Jazz Abdoel, I love sculpture like this, it inspires me to take on more zbrush projects. The take is simple yet so interesting and effective, I also love the colourless finish on the piece. Though I don’t want to compare my own work to something of this quality, it is very evident that I would admire this style of art when comparing it to my own sculpture.



I admired the digital paintings of Eduardo Garcia and Gerard Dunleavy very much, the technique is outstanding, and their command of colour is so commendable. I was intimidated by digital painting before embarking on this project. I found that in digital painting however a lot could be achieved with remarkably little, with understanding and good visual references it becomes easier to produce a painting of quality.

My confidence is in drawing and lines, but I found that if I thought of painting as drawing in its own right, I would overcome any notion of “I don’t know how to paint”. I think this improved my confidence in the area.



Evan Amuldson is a Swedish concept artist and illustrator, as a personal project to maintain his high skill level in design and fantasy illustration he sets himself a “daily drawing” challenge. The outcomes are great and would be considered finished illustration pieces for many artists, due to the high quality.


I really enjoy the work of Philip Harris-George probably because it draws on symmetry, graphic shapes and iconography and the human figure. It’s very interesting, and informed a lot of my early pieces in this concept art project.


Yuri Shwedoff has been a massively inspirational artist for me, I feel like I need a standard to set my work to attempt to match, for future goals. And for me, that is this artist. Absolutely incredible and inspiring paintings, they are peaceful yet unnatural, hinting at violence though subtle, and distinctly surreal. His understanding of setting the particular foggy atmosphere is flawless, and his understanding of composition and perspective, as well as uniqueness of concept is so inspiring to me. I find myself more and more influenced by this artist as I develop my style, my latest paintings perhaps a bit too much. It is important to remember to draw influence from multiple artists and sources rather than just attempting to become one artist, because you will only ever be a lesser version of that artist.



Nivelli 5. Modelling Mouth Shapes

After our mouth shape diagrams had received approval from Johnny at Flickerpix, we were able to get stuck into producing the different mouth shapes. Originally I thought we may be using Zbrush in order to produce the different shapes, but it proved to be more trouble than it was worth, when we could save time producing the shapes in Maya.


We were each assigned different face shapes and split the modelling amongst ourselves. Not all the shapes required teeth, though for the Ernst mouth shapes I chose to produce four that all used both teeth and tongue. This was a difficult task which I had undertaken, there were difficulties with implementing the tongue so that it would print correctly so these shapes required modification. Though because I had already modeled the teeth I thought that I could practice further with using the same face extrusion technique with the tongue.


Kristina had modeled a version of the Ernst face so that we could test the face shapes and see how they worked as almost a 3D prototype. This was necessary so that when we would be printing the final mouth shapes it wouldn’t be blindly and we would have a solid idea of the end printed result.

The mouth shapes I had taken (respectfully, starting top left ending lower right) were “L,TH”, “EH”, “Small AH” and “Big AH”. It was tricky moving the mouth out from its original place, though I found the sculptural tools in Maya to come very in handy for smoothing out the edge loops and keeping the topology intact and printable. I learned a lot about soft selection, topology and Maya’s sculpting toolkit’s capabilities from this task. Although these faces weren’t perfect, these were the ones initially used and we were able to modify and correct them from here (particularly the big puffy cheeks on “Big AH”, these were modified and made more appealing afterwards.)

“L,TH” topology model and model

“EH” topology model and model


After having some practice with the Ernst models I found that Mr H mouth shapes, mustache, were a lot less challenging. Due to the mustache, some of the subtleties in the shape of the sides of the mouth are covered, meaning that there were fewer mouth shapes to produce for this character. “M,B,P” above was a very simple modification, requiring me only to close the mouth to visually communicate these consonants.


P E R S O N A L  P R O J E C T  .  C O N C E P T  A R T

D i g i t a l  .  P a i n t i n g  &  F i n a l  . C o n c e p t s




D i g i t a l  .  A r t w o r k

Screen Shot 2017-05-15 at 21.34.09

Screen Shot 2017-05-15 at 21.33.52


T r a d i t i o n a l  .  A r t w o r k

Screen Shot 2017-05-15 at 21.38.58

Screen Shot 2017-05-15 at 21.39.1217757568_273268086458801_7121927684861582536_n17202804_260336917751918_357736839937055035_n17191166_260336911085252_1858724698375224662_n



S k e t c h b o o k


S T U D I O   P R O J E C T

P h o t o g r a p h y  .  D o c u m e n t a t i o n



M a y a  .  F a c e   M o d e l s


C o n c e p t  .  A r t


M a t t e  .  P a i n t i n g  .  B a c k g r o u n d


Nivelli 6. Matte Painting


I approached Johnny on the possibility of us creating a matte painting background for the project. This was meant to be used as the background in the final animation test using the printed face shapes, completed by Flickerpix and sent back to us for a final seam removal test.

The possibility of created the background really appealed to me, and I was very excited about it. Johnny asked me to have a read through the script and produce a sketch, I am hoping to improve my technical drawing as a current artistic goal. So I am drawn towards architectural imagery or at least an image that challenges and trains my ability to convey structure and perspective.

The image above was derived from scenes that take place within the Frankfurt Hauptbahnhof, I researched and referenced from interiors of the building, I found the interior structure was all very complex. Though not in a convoluted sense, as every form of the structure serves a purpose. It was a very interesting subject to study if nothing else.

Though in following weeks Johnny gave us a different brief to work off of. This was a scene that takes place in a ruined Frankfurt. The architecture has been desecrated and the city is left in ruins, bar the magnificent St Bartholomew’s Cathedral, which has remained intact.

The brief was to produce a digital painting, so I chose to forego traditional sketching and produced a photoshop sketch to communicate how I envisioned the brief. I wanted to maintain something of the exaggerated and unrealistic style established in the other Nivelli backgrounds; while maintaining the structure and the form of the buildings.


We were told to draw inspiration from the work of Austrian-born abstract painter Friedriechsreich Hunderwasser. To achieve this I implemented a style of distortion when drawing the tall cathedral. I wanted to give it something of a supernatural, imposing look, playing to the unnatural circumstances in which it survived the bombing. The idea that I posed to the other two team members working on the painting was to complete a sky that was made up of the line/spiral based designs of Hunderwasser paintings. The overall effect would be something almost psychedelic and perhaps strange visually, but as long as the product we produced would fall into Johnny’s approval, that’s all I can hope for.

Because we were collaborating on the matte painting background we were rotating the painting between ourselves, I completed a large portion of the conception and layout of the image, and also communicated ideas on how it should be coloured, though it hasn’t been returned to me as of yet in terms of rotation.


Nivelli 5. Mocha Seam Removal

Screen Shot 2017-05-15 at 12.26.29Screen Shot 2017-05-15 at 12.27.11

Rotoscoping in AfterEffects is very labour intensive and time consuming, it is difficult to learn the right approach when experiments in seam removal take so long. What I wanted to try for this project was the extension for AfterEffecrs “Mocha”. Mocha is a fantastic program for 3D tracking on footage. Using the same footage from before, the Ernst head turn, I began to track the seam area. The result of the image track is so impressive, and it was able to track the image in no time with minimal adjustments. Very efficient program, one that I have many ideas of how to use in the future for other projects.

To track the seam, you use the pen tool draw a mask around the seam area. Shown in the recording above, you move the time slider across to make sure that the area is covered by the mask throughout. Sometimes the mask can jump especially with footage as shaky as this, but you can move the mask to its correct position throughout and the program will adjust the track accordingly. It’s a very efficient tool for masking compared to the other labour intensive methods I had attempted before. The results for this track were also much better.

The mask creates a shape path that can be imported into aftereffects. Shown in the example below, once you have the shape path it can be exported on it’s own so you have an accurate cut out of whatever aspect of the footage you were hoping to cut out. There are pre existing tools in aftereffects that perform this function; but in my experience using them this semester they require so much more fine tuning to achieve a similar result to Mocha tracking. Mocha is able to track the fine line between the image and the background, even when working with similar colour palettes.

Once I had finalised the mask I exported it and brought the shape into aftereffects, layering it over the original footage. Although my method of removing the line was not a perfect disguise; it did a much better job than my original attempts. I copied the original footage onto a new layer, increased the scale and keyframed the position so that it would copy the skin tone from another area of the face. I then set this layer to track the alpha matte of the shape I had created in mocha.

It’s definitely still patchy and has major areas for improvement, but it is definitely much better than what I was working with at earlier stages of the project. I have removed the seam in this test again working on one side of the face to allow for a comparison of Ernst with and without seam removal.

I think that Mocha is a brilliant tool and I’m very happy I got a chance to see it’s capabilities. I am very drawn to the interesting result that occurs from rotoscoping, there are possibilities with the masking function of this software to create vector versions of footage. The end result is very striking and can be used to make interesting images, the examples I have used above and below respectively are from A Scanner Darkly by Richard Linklater and Waltz with Bashir by Ari Folman. These films are completely rotoscoped, tackling darker and more mature subject matter than most traditional animation. There is more of a link here in that rotoscoped animation can create a more  realistic portrayal that other animation cannot, which can be use for animated media of a greater maturity.

Image result for waltz with bashir scene

Image result for waltz with bashir scene

Nivelli 4. Seam Removal tests

One of the tasks in the project was to complete a successful “seam removal” on the characters once we had been provided an animation test to work with. In the project brief one stage of production is to, once the 3D prints are finalised, we send them to flickerpix Holywood to be painted. Flickerpix will then complete an animation test that we will use in order to The seam is the line where the face of the puppet and the removable mouth shape connect, which needs to be removed in post production. It was agreed upon in early stages of the project that since I was the only one with knowledge of AfterEffects that I should be the one to work on removing the seam, seeing as others had more knowledge in other areas of production, such as topology or the 3D printing and scanning software.

An example of how seamlessly this line can be removed is in Laika Studios’ Coraline; similar to our character Ernst (puppet shown below) the seam is very visible along the mid section of the face. This is opposed to the option of a full removable face shape where the seam is at the back of face and hidden in the animation.The seam is blended so well in Coraline that you wouldn’t be able to tell that it was there at all, this technique of hiding the seam blends stop motion with CGI and has advanced the medium.


There are different methods I could take to approach the task of removing the seam. I want to keep the software for this task, seeing as it is not the entirety of the project, to AfterEffects. It is also a good opportunity to explore more of the features in AfterEffects in terms of post production, up until this stage I have been using it as primarily an editing tool for video-making.

My first tests were unsuccessful in that they failed on most accounts to hide the seam. I was using the Rotopaint tool in AfterEffects, this seems to be a tool similar to the lasso tool in Photoshop. It would be effective to remove a green screen, however for the purposes of seam removal – removal of a small blemish really, it was ineffective.

Also time consuming for less than stellar results, I had to trace the line frame by frame to create a patch to use to cover it up. I was able to essentially clip a layer by colour-picking a skin tone over the Roto line. Good tool though, can see why it might be used, just not for this purpose.

Johnny Schumann provided us with test footage of the puppet Ernst, he moved the camera approximately 180 degrees around the front of the puppet’s head against a green screen background.

Screen Shot 2017-04-03 at 19.39.03

What I used next was the clone tool in order to create a skin tone patch that would blend more seamlessly with the skin tone. This was tricky within itself but works to a certain extent. With a bit more work it could be made seamlessly, or by animating a colour change over. Seeing as this was a test I experimented with this technique on just half of Ernst’s face, to make it more time efficient, but it gives a decent enough representation of what a seam removal technique could look like when applied with this technique.

Compared to the Roto paint tool the Clone Brush worked much better, allowing me to blend the seam, though there were a few visually interesting distortion issues I had to go through before learning the tricks of animating each stroke of the Clone Brush along the timeline. It was pretty cool, think I could go back to playing with tools like this in the future for different projects.

The result was noticeably better than what I had done with the Roto Paint tool, in individual frames the result seemed fine, though in an animated sequence the colour issues showed through. Perhaps with better blending and a more accurate clone, if I was more familiar with the tool, the outcome would be smoother and we could negate this. I was also testing the clone tool over the model in Maya, just as further practice.

The overall result so far is okay, with further practice I think I could manage to pull something decent off with further use of the Clone tool. Although I have been researching the possibility of using a 3D tracking patch with the plugin for AfterEffects, Mocha. It may be the most efficient way to track the seam and hide it while not being too noticeable in an animation.

Nivelli 3. Designing Mouth Shapes

From other members of the group I was given the task of drawing out concept art for the teeth and mouth shapes. Originally we thought there would be a level of creativity in designing the mouth shapes and particularly, the teeth. Similar to what is seen in Laika’s work, such as the Boxtrolls characters. Teeth have the potential to add a lot of character, and seeing as the setting is rather dystopian I was under the impression that we needed to design interesting looking teeth.


I began by drawing Ernst and his mouth shapes, under guidance from other team members that there should be a certain crookedness and character to them. Though after talking with Johnny later in production he had said that teeth were an afterthought and that he hadn’t considered them at the stage of giving us this project.


 Cartoon style drawings are not what I’d usually do. Though I find that due to my understanding of anatomy and technical drawing they’re easy to adapt to and draw. First initial ideas here on what the mouth shapes could be, largely based off the very exaggerated style seen in Laika films. The exaggeration in design is used so that the vowel shapes are emphasized and read more clearly to the eye during the animation.1900.1280.fin.002._L.0033.jpg

However there was a lot of back and forth with Johnny Schumann (our flickerpix correspondant) on the design we had considered for the character’s teeth. He did not like the crookidness and thought that it was just too much and that we needed to scale it back. I still thought there was a little liberty in design so I tried to take it back though still allowing some kind of character design.


This was still not allowed, which is good because we were getting some form of direction, even if it meant there was a bit of misunderstanding between the two of us. We eventually reached a decision on how the teeth should be designed. In hindsight it seems obvious that the teeth should be as basic and simple as possible. More of a basic tooth shape than anything else, though it took a bit of trial and error in order to get there.


Digital sketches I found where a lot easier to create and didn’t take too much time, so in the future for tasks like this I’ll forego the traditional method of sketching and head straight into digital as it streamlines production and communication.

Another job of mine was to find a way to model the teeth onto the Ernst face shape. I was wary because I didn’t know how much I could modify the topology of the model without damaging it in the printing process. I was told that the topology took a long time to perfect for printing so I really wasn’t sure how much I’d be able to modify the model. I was following the logic that since we were all using Zbrush in the previous semester that we had been placed in this studio project for sculptural reasons. I played with the resting face neutral expression in zbrush, carving it slightly to give it the look of the original puppets, though I see that the Zbrush method for each mouth shape would have been incredibly time-consuming. Requiring retopology for each shape.

Instead of using zbrush and having to retopologise each mouth shape in Maya, we looked into using Maya instead. I started by modelling the teeth using an extrusion method from the polygons in the upper lip, it was all pretty simple and didn’t flag up errors as to whether or not it would be printable. Though the question of this tooth shape ‘floating’ was thrown around a lot within our team discussions, but I think it was more speculation than anything. To make the teeth not ‘float’ we would have had to have had a connecting bridge of polygons, which would negate any negative space needed to convey particular mouth shapes (i.e. those with a wide open mouth that display teeth.)

This model test was asked to be modified also by Johnny’s request. The teeth were too big and featured too much curvature. I presented new rendered tests of teeth that were simplified again (straighter, more standard design,) these got a seal of approval and we were able to move onto the next stage of production.



When depicting anything, whether it is a cartoon or not, it’s a necessity for me to study the real world anatomical basis. I took this approach when designing the phonic shapes of mouths pronouncing particular sounds. There is an anatomical logic to it, and each shape represents a function of sound creation. To the eye, we understand this logic, so accurate mouth shapes are necessary to create an accurate outcome. For the design of the Ernst mouth shapes I took our newfound teeth design and applied it to the basic shapes we would need. We had a discussion with Johnny Schumann face to face in Holywood and had been outlined the basic shapes we should produce.


These shapes received instant approval! Which was refreshing, later on in production after animation tests we would find that we would need to include more shapes to make the speech more convincing. This wasn’t surprising, it’s hard to cover all ground with such a limited number of shapes! (Given the Laika figures.)


I ran a quick animation test under Johnny’s advice to use dialogue from the Simpsons. All a bit cute really.