Skydance Silicon Valley filed 3 patents this quarter across 2 categories: AI & Machine Learning (2) and Game Mechanics (1).
The AI & Machine Learning patents cover machine learning systems for automated camera control in games, including real-time viewpoint selection through neural network scoring and cinematic camera management that predicts player reorientation needs. The Game Mechanics patent describes automatic generation of navigation checkpoints that enable timeline scrubbing in gameplay without requiring manual developer configuration.
Skydance's 2 AI and machine learning patents tackle the persistent challenge of automated cinematography in interactive games. The first system replaces traditional designer-placed camera triggers with a neural network that continuously evaluates and scores potential viewpoints, switching perspectives only when threshold gates confirm a meaningfully better angle exists. The second extends this approach to cinematic sequences, where the system predicts how long individual players need to adapt when camera control schemes change and converts subtle player inputs into weighting factors that influence automated camera selection without handing over direct control.
The single game mechanics patent addresses accessibility through algorithmic checkpoint generation. Rather than requiring developers to manually place save points or navigation markers, the system analyzes narrative structure and gameplay data to automatically identify stable game states that can serve as valid entry points, enabling players to scrub backward and forward through a game's timeline much like seeking through a video file.
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.