ARKit & other face tracking mistakes

Many tech leads live under the assumption that – if they acquire enough data to train their model, problems with quality will simply work themselves out. Wow! Magic. This assumption often operates under an additional (but false) assumption: There is only a negligible percentage of impure data.

Read More...

cheek raiser vs. lid tightener

Whether you’re using an expression model to pose AUs for academic research, product-based machine learning, or character art, you will face challenges acquiring pure examples of cheek raiser and lid tightener.

Read More...