Completely agree. There are three primary forms of bias: Human Bias, Data Bias, and Algorithmic Bias. A better solution is to improve existing machine learning models with existing solutions (e.g., domain adaptation, domain generalization, discovering latent domains). This will improve the overall performance.
Thanks for the feedback. Can you speak more to "null space optimization", "self intersection", and "hybrid position/force control"? High level descriptions of the situations would be helpful, thanks. Or links to resources.
Hey, thanks for reading. There are a TON of ways to implement Inverse Kinematics, and I'm going to cover the most popular ones in coming articles. Do you have any preferred animation techniques you'd like me to cover? All of my articles will be available here: https://medium.com/unity3danimation
I'll make a plug for software I wrote based on sequential quadratic programming. It performs significantly better than inverse Jacobian methods in the presence of high-dof and joint limits. Here's a link to the software and paper: https://bitbucket.org/traclabs/trac_ik
What is the best IK algorithm to implement in JavaScript given that it has to be small enough to download but also robust and fast and probably something one can implement without spending too much time on it? Probably an impossible question... but I figured I'd ask in the hopes there was something easy and obvious here.