18073376. SYSTEM AND METHOD FOR AI SEGMENTATION-BASED REGISTRATION FOR MULTI-FRAME PROCESSING simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
Contents
SYSTEM AND METHOD FOR AI SEGMENTATION-BASED REGISTRATION FOR MULTI-FRAME PROCESSING
Organization Name
Inventor(s)
Akira Osamoto of Plano TX (US)
John Seokjun Lee of Allen TX (US)
Hamid R. Sheikh of Allen TX (US)
SYSTEM AND METHOD FOR AI SEGMENTATION-BASED REGISTRATION FOR MULTI-FRAME PROCESSING - A simplified explanation of the abstract
This abstract first appeared for US patent application 18073376 titled 'SYSTEM AND METHOD FOR AI SEGMENTATION-BASED REGISTRATION FOR MULTI-FRAME PROCESSING
Simplified Explanation:
The method described in the patent application involves separating foreground and background in a scene by generating a segmentation mask using a reference frame. This segmentation mask is then applied to multiple image frames to create foreground and background image frames, which are further processed through multi-frame registration to generate a combined registered multi-frame image of the scene.
- Obtaining a reference frame from multiple image frames of a scene
- Generating a segmentation mask to separate foreground and background
- Applying the segmentation mask to create foreground and background image frames
- Performing multi-frame registration on foreground and background image frames
- Combining registered foreground and background image frames to generate a combined registered multi-frame image
Potential Applications: This technology can be used in video editing, surveillance systems, augmented reality applications, and medical imaging.
Problems Solved: This technology addresses the challenge of accurately separating foreground and background in complex scenes.
Benefits: The method improves the quality of image segmentation and registration, leading to more accurate and visually appealing results.
Commercial Applications: Potential commercial applications include video production, security systems, virtual reality experiences, and medical diagnostics.
Prior Art: Prior art related to this technology may include image processing algorithms, video editing software, and computer vision techniques.
Frequently Updated Research: Research on improving segmentation accuracy, registration algorithms, and real-time processing capabilities may be relevant to this technology.
Questions about the Technology:
- Question 1: How does this technology compare to existing methods of image segmentation and registration?
- Question 2: What are the computational requirements for implementing this method in real-time applications?
Original Abstract Submitted
A method includes obtaining a reference frame from among multiple image frames of a scene. The method also includes generating a segmentation mask using the reference frame, where the segmentation mask contains information for separation of foreground and background in the scene. The method further includes applying the segmentation mask to each of the multiple image frames to generate foreground image frames and background image frames. The method also includes performing multi-frame registration on each of the foreground image frames to generate registered foreground image frames. The method further includes performing multi-frame registration on each of the background image frames to generate registered background image frames. In addition, the method includes combining the registered foreground image frames and the registered background image frames to generate a combined registered multi-frame image of the scene.
