Adaptive object detection and recognition

This invention describes a method for analyzing video to identify objects. It first uses a lower-resolution video to quickly find objects, then switches to a higher-resolution video to recognize them in more detail. The system can then use feedback from these steps to automatically adjust the resolution of either video stream, aiming to improve its accuracy in detecting or recognizing objects. The claims specifically detail adjusting the resolution of video frames based on the success or failure of the detection or recognition process.

Why it matters: In 2018, real-time, high-accuracy object detection and recognition were computationally intensive. Since then, advancements in deep learning models and more powerful, accessible AI hardware have made such adaptive video processing systems significantly more efficient and effective, allowing for dynamic resolution adjustments with better performance.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedSoftware logic, machine learning, video processing

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10