Adeia filed 5 patents this quarter across 3 categories: Streaming (2), Cloud Gaming (2), and AI & Machine Learning (1).
The Streaming patents address peer-to-peer game Streaming with congestion-aware scheduling and adaptive quantization parameter control to reduce latency during scene changes. Cloud_gaming applications focus on automated camera angle switching using virtual IMU data to prevent motion sickness in remote gameplay and spectating scenarios. The AI & Machine Learning patent covers an in-game coaching system that imports skilled player models to deliver personalized, real-time gameplay advice.
Two Streaming patents tackle the challenge of delivering live gameplay to multiple viewers with varying network conditions. The first describes a peer-to-peer system where a game console acts as an RTP server, using the slowest client's feedback to pace all transmissions and automatically downgrading struggling viewers to a spectator mode with optional AI-controlled stand-ins for their characters. The second addresses motion sickness in streamed gameplay by embedding virtual sensor data derived from in-game camera movements directly into the video stream, allowing each viewer's device to automatically switch camera angles based on personally calibrated motion tolerance thresholds.
Cloud gaming applications filed this quarter focus on latency and comfort during remote play. One patent prevents the cascading frame drops that occur when sudden scene changes spike bitrate demands, applying preemptive quality adjustments before a stall forces the system to send large recovery frames that themselves risk being dropped. The other mirrors the Streaming category's approach to motion sickness prevention, using virtual IMU parameters computed from game engine camera data rather than physical sensors to trigger real-time camera switches tailored to each player's sensitivity without requiring manual adjustments.
A single AI & Machine Learning patent describes a coaching system that brings expert guidance into active gameplay sessions. Instead of replaying recorded tutorials or offering generic hints, the system loads a trained model of a skilled player's decision-making patterns and uses it to generate contextual advice as the current player progresses through a level or match, with the option to synthesize that guidance in the expert's voice for added authenticity.
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.