I have a bear mesh that is the combination of meshes from different files/different “versions” of this bear. Some parts of the mesh are disconnected/overlapping.

I have tried different armatures on this mesh and am now trying to use a “basic human meta-rig” to control the mesh. My questions are about how vertex groups in this mesh are working. In the “vertex groups” section, I do not know how all of the groups got there: are they from past armatures related to the same mesh? Are they from libraries that are shared across different meshes/files?

The basic human meta-rig did a good job controlling the mesh but had a few weight-painting issues. Because this vertex groups list was disorganized and I could not quickly navigate through it for weight painting I decided to delete all vertex groups and re-parent the basic human meta-rig. (However, it seems to me that entering weight paint mode with the armature selected, then shift-clicking bones, is a more efficient way to do this - do you agree?)

After deleting all vertex groups and re-parenting the basic human meta-rig, it did a much worse job controlling the mesh. I needed to go back to using the mesh with its disorganized list of vertex groups to get the meta-rig to work again. This made me think that somehow the meta-rig was working with previous vertex groups. Is this correct, or am I missing something?

Finally, I tried a different version where I used a basic human meta-rig and corrected all weight painting issues. In pose mode, the armature controlled the mesh exactly how I wanted it. Then I clicked “generate rig” and parented the mesh to the rig, went to pose mode, and found that the rig’s control didn’t correspond to the meta-rig armature’s control. Why is this?

In general, do you think that using a meta-rig the most efficient way to control a mesh like this? I would like to be able to easily animate simple movement - walk/run, dance, sit, swim, etc. I will appreciate any insights on how to better animate and understand these concepts. Thank you.



1 Answer 1


You should follow some tutorials on rigging, like Darrin Lile's or Sebastian Lague's, it's better to first learn from scratch in my opinion.

Before parenting it's always better to make sure that the transforms (scale, rotation) have been applied, and that the normals of your mesh are correctly flipped.

For the vertex groups, when you'll parent the mesh to the armature With Automatic Weight, it will automatically create as many groups as you have bones, and the groups will have the names of the bones that control them. If the mesh already has some vertex groups from previous operations, it will keep them as long as their names are different but it won't interact with the parenting, so you should not have to care about that but you can always delete them if you prefer.

If you parent With Empty Groups, it will create as many groups as you have bones, but it won't affect any vertex to these groups, you'll need to do it either with the Assign button or in Weight Paint mode, this is the best method when you parent non organic objects like machine, robot, etc. as it let you define what part is supposed to be controlled by which bone and doesn't create any gradient influence.

For the armature, you can use the ones you currently have, except it seems incomplete, you should have a root bone, your lower legs or lower arms should have IK constraints, you should have the targets and pole targets for these constraints (you already have targets for your legs though), and make sure that these bones (root, targets, pole targets) have their Deform options disabled (which is not the case for your leg targets).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .