Camera Preview Object Removal via Preliminary Action
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capture technologies do not allow users to preview an image with unwanted objects removed in real time, forcing users to capture multiple images and remove objects post-processing, which is inefficient for pre-capture adjustments.
Innovation Solution
A method and device that receive multiple images from an image sensor, classify pixels as moving or background, and generate a real-time or near-real-time preview image excluding unwanted objects, allowing users to make adjustments before capturing the final image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If multiple pictures are captured and synthesized in post-processing to remove unwanted objects, then the unwanted objects are removed from the final picture, but the user cannot see the clean picture in real-time and must make adjustments after capture
Solution Approach 1:
The system performs object removal in advance by generating a clean preview image from multiple captured frames before the user makes final capture decisions. This allows users to see the effect of object removal and make framing, focus, and composition adjustments based on the cleaned preview, rather than discovering object removal only after post-processing.
Solution Approach 2:
The system creates a copy of the scene from multiple captured frames and processes this copy to remove unwanted objects, displaying the cleaned copy as a preview while preserving the original captured frames for final image generation. This allows real-time preview without compromising the quality of the final capture.
2Ease of operation
If real-time preview with object removal is generated, then users can make better pre-capture adjustments, but computational complexity and processing requirements increase
Solution Approach 1:
The system applies object removal processing selectively to generate previews at lower resolutions or for specific regions of interest, rather than processing entire high-resolution images in real-time. This reduces computational burden while still providing users with sufficient visual information to make meaningful pre-capture adjustments.
Solution Approach 2:
The system performs preliminary processing by capturing multiple frames and identifying moving objects in advance, storing this information for use in generating clean previews. This preliminary analysis reduces the real-time processing burden during preview generation.
3Reliability
If multiple images are captured and stored for synthesis, then unwanted objects can be removed effectively, but memory and storage requirements increase
Solution Approach 1:
The system extracts only the essential information from captured frames—specifically, the identification of moving objects and their trajectories—rather than storing complete high-resolution copies of all frames. This extracted information is sufficient for generating clean previews and final images while significantly reducing memory requirements.
Solution Approach 2:
The system uses temporary low-resolution buffer frames for real-time preview generation, which are discarded after use, rather than maintaining large buffers of high-resolution frames. This approach provides the necessary functionality for object removal while minimizing persistent memory usage.
Data Source
AI summary
An electronic device receives a plurality of images at different times while the device is operating in an image preview mode, and prior to a user-initiated image capture request, the electronic device generates a clean preview image that excludes one or more moving objects from the preview image.