Disney Enterprises received 1 granted patent in AI & Machine Learning.
The patent covers technology for training virtual character motion controllers using signals from game controllers, addressing the disconnect between training data and real-time player input. This approach enables smoother and more natural character animation by directly incorporating the types of inputs players actually use during gameplay.
Disney's single AI and machine learning patent tackles a persistent challenge in video game animation: the mismatch between how neural networks are trained and how they operate during actual gameplay. The system trains motion controllers by extracting features directly from game controller signals and matching them to pre-recorded motion sequences, ensuring the network encounters the same kind of data during training that it will see when players take control. This eliminates domain shift, the phenomenon where a model performs poorly on real-world data that differs from its training set, without requiring additional adaptation layers or extensive re-labeling of existing motion capture libraries. The result is character animation that responds more naturally to player input because the underlying model was trained on the same type of control signals it processes in real time.
All data sourced from USPTO patent filings. Google Patents may take several weeks to index recent publications. If a link is unavailable, search for the patent number at USPTO Patent Public Search.