This video explains what it is doing in box modeling process with an actual example. It also explains how we can use two views of the object to form a 3D model from them.
Here is the link to the tutorial for Box modeling a Car with Quad Topography. It is for a free-form generalised car model.
Rhino 6 subdivision modeling- simple cup.
This is a tutorial to modeling mesh objects in Rhinoceros 3D. I find it interesting because Rhino is in fact a NURBS modeling software but supports also meshes, and users can translate the models in between different representations effortlessly. For example, this (http://www.studiorola.com/tutorials/3d-modeling/rhino-tutorial-nurbs-from-subdivision-model/) link provides a tutorial to translate meshes into NURBS poly surfaces. Each representation has their benefits. For example, meshes are east to edit, but NURBS surfaces are preferred in CAD/CAM for its precision.
Here's a goofy lil video where they model a simple alligator from a sketch!
The beginning is a little fast, since they Catmull-Clark subdivide the cube right off the bat and start sculpting the tail, but besides that it's pretty easy to follow. The only part that stuck out to me was towards the end (around the 7:15 mark) the artist uses some sort of collective shrinking tool that lets them deform and reduce the diameter of an entire slice of the alligator's neck. That and the artist's good use of pulling faces out make this a good beginner video that ends up with a simple but good looking end result. Besides the fact that the artist skips in the last 5-10 seconds to an alligator model with some accents like eyes and nostrils and teeth added. :)
Box modeling a Pig : https://www.youtube.com/watch?v=-AxbFFS4MS8
At 9:34, the creator points out pointy surfaces such as the nose and feet. But then he adds a loop which I didn't quite understand how it functions. Anyone could give me an insight on that functionality? The rest of the modeling is simply done using extrude and simple element bevels with subdivision. Also, at the very end i.e 13:00 the creator creates a sphere for modeling the eye and moves it to the eye socket. My question is, when we export such a model, does it save the sphere differently than the remaining body of the pig or does the sphere somehow develops some connectivity with the remaining mesh?
This video is a super sped-up clip of someone modeling a house in auto-desk basically from scratch: https://www.youtube.com/watch?v=U_RM3coUgx4
Most of the manipulations were straightforward things like beveling, extruding, and translating vertices/faces. The coolest thing I noticed was that, towards the end of the video, the artist pulls a whole neighborhood of polygons outwards, with the vertices closest to the 'grabbed' point being pulled the farthest along. Another interesting technique they used was constantly switching to a subdivided view of the mesh for a glimpse at the final product: I wonder if this was implemented by actually subdividing and undividing (?) the mesh whenever they press the button, or perhaps by instead maintaining a separate, subdivided view of the mesh.
This video shows a smart strategy that transfers a 2D image to a kind of "3D" model, which is a better reference for box modeling.
This is a video box modeling Hiroshi from Big Hero 6! One thing I thought was interesting about this video was how the person chose the vertexes. By that I mean at 1:51 the vertex that looks like it would represent the nose is much higher and not actually at the nose. It made me wonder about how people know what to prioritize first, and when things can be accounted for later rather than early on.
Most of the spaceship mesh this artist forms is done by box modelling. They use a variety of local mesh operations including bevels, edge translation, face deletion, etc. They also use subdivision to increase the detail of the mesh.
Modelling a vintage car. Its very interesting how the interior of the car is dug out of the body rather than the rest of the body being built up around it, while many of the features that protrude from the car are made using the usual bevel method.
It turns out box modeling can be useful even for creating spheres! In fact, it seems to be more preferable in some cases (compared to starting with a default sphere) since it creates less faces, and the faces it creates are more even in size. The faces are also all quads, whereas a built-in sphere usually contains both quads and triangles since it is divided into faces along longitudes and latitudes.
Here's a demo video showing this difference:
a video of boxmodelling a creature. I mostly find it interesting how the artist "drags" new faces out of previous ones and uses those to create limbs in a way seemingly similar to that of creating curves with control points and tangents. I also thought that it was neat how the artist made just one half of the model and copied over the other side with a symmetry plane.
This tutorial covers nearly everything I needed to know about using Maya (definitely including some cool box modeling tricks). VERY HELPFUL! Sign in by Organization -> cmu.edu
I think this video is really cool. It derives the model of human body from three projections and then processing the box modeling.
This tutorial shows how to make a low poly box and barrel in Blender just in 10 min. The models are simple and they look nice. I like the way they add depth along edges to give the model good details.
This is the tutorial to the subdivision of a mouse in Blender. It also illustrates how subdivision works, both on NURBS and Quads. The final geometry is legit!
This tutorial shows how to start out making a cartoon character's face in Blender using extrusions and subdivision. If you find it interesting, he has more videos that show him finishing the rest of the model.
The final tree model might be a bit rough for today's graphical standards, but I thought it was really interesting how the trunk and branches were shaped by distortion along splines; I had not seen this method before. It was also cool to see how the author worked without a fixed reference.
Here's a cool tutorial using Maya. They use box modeling to make... a box. But it's a really cool, very textured box made using some tools that are not regularly part of Maya.
At one point in the video (around 5:20), they "increase subdivisions", which seems to soften the corners of the box without actually turning the box into a rounded shape. I have a guess about how that works: by increasing the number of subdivisions, the part of the box getting rounded in the subdivision surface is smaller, so only the faces near the corner are rounded, rather than the whole face. Is that right?
And, in general, what can one do when they want a shape to be partially round and partially sharp?
Here is a really cool video of the box modeling techniques for Mario. I find this super interesting because you can see how 3D modeling needs to changed for memory and for the modeling power of the GPU itself. In early years, game designers had to figure out designing with low polygon count. A few years later, some better designs arose using better textures. Notice how animators attempted to solve some of the issues by creating multiple masks for the character instead of morphing the body itself. Finally, in recent years, high polygon count finally kept up and the modeling of these 3D characters became easy for animations and character movement.
I watched a video about character modeling using box modeling that follows many of the same concepts as used in class where an image is used in a few directions to really feel out the basic geometry of the character. What interested me was the realization of how frequently the mirroring of characters is used to achieve a symmetric body even though the image itself didn't have the same symmetry. We as humans don't inherently have perfect symmetry, but I think because our minds would like to see it it doesn't come as off-putting that most of our art has it even if it isn't fully realistic.
In this video, they model a human face from sketch. It seems more advanced than other videos I've seen, where the artist would use a 2D sketch in the background to do one side at a time, unlike in this video where the artist does all face simultaneously without a sketch.
Moreover, there are very few face bevels involves, with most operations being either edge bevels or scaling.
This is an impressive demonstration of what artists could do with box modeling. The author hand crafted a Benz car completely out of primitive base mesh, referencing three views of the car. One thing I learned from this tutorial is the importance of checking reflection and shading effect frequently, so that the final render with lighting effect on would look nice and smooth just like a real world surface.
An interesting video in which box modeling is used to create a model of Pikachu. I found a couple of the decisions that they made when modeling the ears to be interesting: first, they created a new mesh for the ear, rather than extending the mesh used for the rest of the model; and second, they spent a surprising amount of time looking at the control cage when editing it, even though a couple of the points at the tip were nowhere near the subdivision surface. I'd guess that the former is because they didn't want the interface between the ear and the rest of the head subdivided (preferring the sharp edges there), though I have no idea what made the control cage a more useful view when editing the diameter of various parts of the ear.
This is an interesting link which demonstrates how adding/deleting edges results in varied subdivision models (smooth models) while you may start with the same initial geometry. This is a good insight in understanding that subdivision algorithms are mesh dependent and final results may vary based on the quality and type of initial mesh chosen by the user.
A video of box modeling character Hiro Hamada from big hero 6. I found the process in making the hair pretty interesting, they are just adding leaf shaped objects and overlaying it on the head. I wonder if the actual hair has any volume, whether there would be a good series of transformation to add volume if the hair was just planes.
Here's a video of somebody animating a kid's head in Maya. I thought it was interesting that the creator only designed one side of the head and then duplicated it to have a nice symmetrical face. One of my friends from high school who does a lot of art on the side occasionally messes up a face because he can't copy and paste the other side of the head but has to draw it all again in pencil.
A video of a Blender tutorial showing how to box model a rabbit's head. Actually, when I took 15-112, my term project involved some Blender usage. To make myself more familiar with Blender, I actually worked through a few tutorials including this one! I found it interesting how the nose and ears are still extruded from the original box even though they do not resemble a cube at all.
This is a video of box modelling a simple character. Not sure how to stitch together various meshes for the body parts..
Here's a video of box modeling of a hand. In this video, box model is used to first create the palm of the hand and just one finger. Using subdivision, intrinsic details are details to the model. Then fine model of one finger is used to model other fingers as well. One thing I learned from this video is that how we can use multiple mesh models and join them together by matching the nodes at the intersection as is done with the palm and fingers in this video.
This is a really good overview of how subdivision modeling works as a workflow and it shows the reasons for best practices such as only using quads instead of triangles and the justification behind those guidelines
Relatively quick tutorial that goes over all of the basic box modeling techniques. Includes clear demonstrations of all the techniques need to create the models for the quiz (extrude, scale, bevel, etc.) which I found very helpful.
A very good tutorial of box modeling. It shows how to model a cup from a box. Also a little bit about the light and material render.
This shows a naive but intuitive hand modeling process
This tutorial shows box modelling of an ear. Instead of starting from a box, it started with a thin layer, found those mesh points and adjusted where they should go by modelling the ear in three dimensions. I am wondering if this is another type of box-modelling as the one we saw in the class started out with a box.
https://www.youtube.com/watch?v=Wl-f5p0GdP0&t (until about 2:30 for actual box modeling)
I thought this artist used a lot of interesting tricks to sculpt a rhino. In particular, I noticed that they tend to carve out inner body details like the mouth or the inner ear. Also they tend to use something like Edge Split or like the opposite of Edge Collapse to pull out additional vertices from existing ones and add detail and dimension.
In general, I notice that a lot of artists actively avoid using anything but quad shapes in their modeling (i.e. when people model spheres they subdivide and smooth a box as opposed to going for the default sphere, which has triangles converging at a point). Why? Is it specifically to make various subdivision properties easier or is there something later in the pipeline that isn't conducive to non-quads?
The tutorial shows how to effectively construct a smooth mesh with orthographic images. A lot of tricks it does, for example extending edges and splitting edges, are general operations required in mesh editors of all kind.
This video shows the box modeling of a head. I thought it was really interesting how complicated it looked just to make a head, which is mostly round already. In particular, the nose and lips of the head seemed to take a lot of steps.
Modeling of an ear, ear is first subdivided into rigid components and directly transferred into 3dsmax. It's also interesting how the author's approach to modeling the depth of the ear; he's also using all three views to make sure there are no leaks on the surface.
This tutorial shows how to box model a cartoon head. Lots of basic mesh edit functionalities are used and well-demonstrated. So it is a very good starting point for guys like me who doesn't have much experience in mesh editing before.
It's a tutorial of modeling with T-splines for Rhino3D. The method is quite similar to Scotty3D and one interesting thing I noticed is that user can edit vertexes, edges, and faces in both control cage mode and subdivision surface mode. And in the latter mode, the points/edges/faces are the ones in control cage mode being projected on the subdivision surface.
This is a just a video of someone box modelling a simple human head starting with a cube. I think it's really cool how so many tiny details go into something like this. At 1:08, when they grab the entire nose and mouth and move it around, it really gives you a sense of how precise we expect faces to look in real life and how strange they can look if they're a little bit off.
This is a nice tutorial which allows you to clearly see step by step what is going on in the process of modeling a car from a cube. From the step by step images, it is easy to work out what was done in each step. I am somewhat unsure as to how the wheels are added into the model.
This tutorial shows how to box model a character (presumably for a game). I like it because the pace is slow and the video is detailed, so it's easy to get an understanding of what is going on and how it could be replicated/similar things could be made by someone without a lot of experience. I felt like I learned the basic steps I could use to create my own character using box modeling.
This shows the creation of a hand from a box. The thing I found the most interesting was the creation of all the fingernails around 22:00. It really showed how subdividing lets an artist create really fine details on the final model. As a whole I also found this tutorial really showed off the process well of this style of modeling.
This is a video on the creation of Anna from Frozen. First of all, modeling a character like Anna takes a lot of time, focus, and a strong eye for detail. Even with the tools. However this video only models her head. When modeling a character do you typically model it in parts i.e. arm, head, body and stick the parts together, or do you model they entire character entity and then animate that?
This video explains the use of blender a popular graphics tool