Introduction

This here is my individual blog for the Industry Exercise assessment. My group and I had developed a project named "Deep Sea Encounter" and to produce a short animated trailer for the storyline. Here is the progress in which the project is being accomplished.

Working Title – A Deep Sea Encounter

Synopsis - A marine biologist and his eleven-year-old son embark on a journey of discovery in a submersible pod in the deep sea where they encounter a strange creature unlike anything they’ve ever seen before.

Plot - A marine biologist, Dr John Brantwood, and his son, Andrew, go on a journey in a submersible pod to the bottom of the sea, miles below the surface to discover any creatures that they have not seen before. Upon discovering an unknown cave that is filled with bioluminescent plant life, their pod accidentally loses power through physical damage. Stranded in a cave and unable to call for help and a certain amount of oxygen left, the boy discovers an unknown creature: a female looking humanoid sea creature with bioluminescence. Upon encounter, the two develop a bond and the fish girl summons a group of creatures that gather to get the pod out of the cave and they produce enough light to attract the attention of a rescue team who has come to retrieve the pod and the passengers after the system powered down.

Time – 1990s, daytime, summer.

Characters – John Brantwood, a marine biologist who specializes in deep-sea excavation. He is hardworking but he attempts to spend as much time as he can with his son.

Andrew Brantwood – John’s eleven-year-old son who has isolated himself from other children ever since his mother left. He has a thirst for knowledge, especially in marine life having been inspired by his father.

The Deep Sea Angel – an unknown bioluminescent creature that shows a level of intelligence when she summons some of her kind to assist the humans after their pod is damaged.

Theme – Discovery, dark, mysterious, alien like, adventure. Light to dark.

Concept Work and Development

Below are a series of images of concept artwork and illustrations that were to be used in the unfinished novel. Most are hand drawn and scanned into the computer where they were coloured and textured using Photoshop. Attempted to use certain colours to signify certain moods and themes in the story.

 Original concept image of the submersible pod descending into water.

Another design for the setting of the story. 

A scene occuring after the submersible pod has crashed and been stranded in the original story. Achieved the lighting effects using Photoshop, used red mostly for the theme of danger.

Another illustration that highlights the interaction and relationship between the young boy and the jelly maid. Attempted to use warm and cool colours and used other textures as well as other effects like the boy's reflection in the glass window.

Original layout sketch for the "jelly maid" to be used for reference for modelling in Maya. Only half of the creature was drawn as I intended to mirror the image later using Photoshop.

Same image after being cleaned and manipulated in Photoshop.

Side view sketch of the "jelly maid." Made the proportions as accurate as possible with the last sketch, the arms aren't drawn in, just gave a rough position of where the shoulders would be. 

Simple sketches of how the jelly maid's hands would look and be constructed.

Diagram of the anatomy of the jelly maid, created while conducting research on different marine creatures, just to add a bit of logic to the subject.

Final Storyboards & Animatic.

Once a script or an idea is brainstormed, one crucial method to making it visual is by producing storyboards, comic book like forms of the story. Below are images of the final versions of the storyboards, based on the orginal ideas that we had for the shots.

For the animatic, we did go through a series of ideas before we settled on the final version. Below are two animatics, one was done by another member of the group while the other was done by me, using images from the storyboards shown above. No audio has been done for any of these but will include them later on.

This animatic was done by the rigger of the group, once other modellers in the group had constructed a rough version of the environment, he created a layout of how the shots would look from different angles and the right pacing.

This animatic was produced by me, using images from the final storyboard. I did not want the finished version of the trailer to be too long, so I was trying to aim for arouhnd 1 minute running time, although with the kind of shots we are producing, we may need it to be longer.

Work in Progress - Characters - "Andrew"

This post shows the progress of developing one of the main characters for the original storyline. As the story involves a young boy, I produced some concept work of the boy including illustrations to use for reference to model the character later on in Maya.

Concept Artwork

Modelling/Texturing

In order to bulid the character, I had to import and use some concept images for reference.To do that, I simply created a polygonal plane, turned down the subdivision surface modes and scaled it up to a decent size. Then I assigned a new lambert material to the surface of the plane and imported a JPEG photo file of the image that showed the body structure of the character.

Only one half of the body was produced at first using a single polygonal cube, along with the arm. Then I composed the leg with another cube. Once all boday parts were complete, I combined them into one mesh, then merged certain vertices to others in order to secure the model even further. With construction history deleted (doing this will prevent problems with the mesh in the future), I mirrored the object to create the other half of the body, but not before aligning the vertices of half of the body in order to avoid complications.

The image above shows all the vertices of the open side of the mesh being snap-aligned. It is important to do thise every single time you wish to mirror the geometry of the mesh to create the other half as not doing so would cause problems. If for any reason, the mirror geometry mode fails, I would have to mirror the geometry with the mode to merge vertices with the original switched off and then go through the process of merging appropriate vertices together manually.

The shirt, trousers and shoes that I have produced are actually created from duplicates of the original mesh so they would all fit on the character with a few modifications. Later on, I would also create a lifejacket for the character as it would be relevant to the setting of the story.

Son character in low resolution. Some of the geometry/topology may need some cleaning up to do but it is looking pretty good. The shirt has assigned a new lambery material and given a new colour so as not to get confused with either one of the objects.

A screenshot of a smooth mesh version of the son character. Applied simple lambert materials to each object and a phong material for the eyeballs. The area around the ears on the character are not perfect but that could be solved with other models in the future.

Screenshot of the son character with some textures applied. Started creating and adding hair using CV curves and a hair system. Textures for eyeballs were creating using Photoshop files that I produced years ago and applied using a spherical mapping technique. With the original mesh, I applied automatic mapping to the surface, and then went to the UV Texture Editor to take a snapshot of the mesh's outlines. Then imported it into Photoshop and produced a good skin texture with some freckles around the face area. Saved it as a JPEG file and then imported and assigned it to the colour section of the lambert material.

Another screenshot with a rendered image of the character's head, with some simple textures and hair.

Wireframe mode screenshot of the smooth mesh with the hair, CV curves and other body part objects included.

A screenshot of the boy charcter near completion.

A close up of the character with the main mesh outlined in object mode.

Character Development - Father

The father character was produced separately, the head was created by another member of my group and I produced the body and the clothing to go with the character. The geometry was quite difficult for this character as there were too many edges and loops in the head's topology, but I managed to clean up a little of it in order to successfully combine it with the body mesh, merging appropriate vertices.

The complete meshes in wireframe mode, after they were constructed using reference images.

An image of the head, with one polygonal sphere for the eyeball and a simple face of the mesh selected.

Father character in smooth shaded mode. Attempted to produce textures for different surfaces such as the mesh and the "eyeballs." Accomplished it by assigning image files to existing materials, which were of course assigned to the specific surface. Once the modelling and texturing was complete, the character was sent off to be rigged.

Production - Newspaper Shot

In the original storyboards, we decided to include a close up shot of a newspaper revealing. This was to show and tell the audience more of the story plot that is about to take place.

I composed the newspaper out of a polygonal plane, then I manipulated part of the mesh, folding it over so it would look more like a newspaper. Attempted tests with the assigned texture and lighting and shadow in the scene, but in one test, the edge of the newspaper where it was folded was out of focus, which caused a slight problem in the texture material. To solve the problem, I added more edge loops to the area.

Next was to make the mesh move like paper, and to do that, I applied an nCloth mode to the object. Of course, I had to create another plane for the floor so that the newspaper object would collide with it. Simply made it a passive collider and then altered the settings so it would all collide and no part of the mesh would pass through it. Then I altered the controls of the nCLoth asset such as gravity, wind, stretch resistance, so that the object would move realistically like paper. The image below shows part of the settings.

The texture image was produced using Photoshop. First, I did a bit of research by reading newspapers and the internet for reference on producing an authentic looking newspaper. To produce a good quality image, I increased the resolution of the file so everything would show clear and sharp and not be too blurred.

I produced a template of the newspaper by using shapes and text boxes, making sure the template was even for both the front and back side of the newspaper. Then I began writing titles and a paragraph of the story. For the picture, I decided to use an early illustration I produced for the story. I saved the current image for a test shot, then when it came to rendering the final version of the shot, I decided to make the newspaper texture look old and dirty. To make a bump map, I created a file of the same size and resolution as the image itself, then make a crippled paper sort of texture by using the setting "Difference Clouds" and "Emboss." Then I changed the levels so the texture would show up more and then saved it as a separate file.

The image above is the finished texture file of the newspaper. Made it look more worn, stained and old by using different brushes presets and overlay setting.

An image of the original camera shot with the hardware texturing and the nCloth mode active.

Final rendered image.



Sound Effects

Of course, we would need audio to accompany the visual, so that's why we had to find some sounds and music. We presented our idea to a group of people who specialized in the sound department and a few of them volunteered to produce some musical compostions. We sent them an animatic to get a feel for what sort of theme and style would suit the setting of the story. Sometime later, they sent me a few audio files of some quite interesting themes that would be good. We managed to use one later on for the final trailer.

For other sound effects, I set on doing some research and found some sounds that were relevant to the environments that were involved in the original plot. For the ocean, I found some sounds that included a slight breeze and waves gently crashing. For the pod being lowered into the water, I had to find a sound of a splash that had weight and I managed to find one. Of course not all these sounds were right in some way, which is why I used Adobe Soundbooth to manipulate them to our liking.

For several sounds that involved underwater bubbles, I gave them a resonant effect and a reverb to make them sound slighly echoey.

In this image, I am attempting to create another resonant effect for a sound effect that is to be used for one of the underwater scenes.

Once all of the sound effects were done and converted into WAV files, I imported them into Premiere Pro to arrange them with certain scenes. In the image above, you could see the pod being moved by a crane, so I used a combination of sounds including creaking metal, whirring and quiet ocean for the background.

Final Movie Link & Review

The entire sequence of shots was edited by the co-member who produced textures for most of the shots while most of the audio was done by me. We went through a session of deciding which shot went where, how long each shot had to be whether it was too long or short and where we needed titles.

Below is the link to the finished movie clip.

FINAL DEEP SEA - YouTube

Now that we have come to the end of producing the final piece, it is now time for a well earned rest and a small celebration, but not before giving a little review on the progress of the project. All in all, the project was well planned and we made good sufficient work. There were not too many complications or difficulties on all assets and roles that we were assigned, with the execption of a few technical errors with the software and the render settings. For the finished movie clip, on a critical note, there are certain aspects and scenes that could have been changed or improved. For the visual side, there are noticeably a few continuity errors, due to using different scenes set up or same scenes changed by different users. The lighting may be slightly over the top in certain areas but in others it has set different moods as the trailer goes. The designs of the environments were really well done and produced, the characters themselves could have been developed more at the beginning of the project, in concept and modelling, but they weren't done too badly either throughout a short period.

As for the audio, the music was done by a volunteer from the sound department, who was able to produce a good composition for the trailer; it may seem quite repetitive though as we had to alter it to an extent in order to make it fit and the other sound effects may be slightly louder than anticipated. Could have asked him to make some changes if we had more time. But other than that, sound and music really do help to set a mood for an animation.

Once again, the project was well done; great work to all my fellow group members who managed to pull it off and I hope we will work again to produce better animations. Practice makes perfect.