Image Merging via Automatic Object Placement and Lighting Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image merging techniques on mobile devices are time-consuming and inefficient, requiring users to perform multiple operations to seamlessly integrate objects from one image into another, often resulting in unnatural or unconvincing composite images.
Innovation Solution
A method and system that analyze lighting conditions and object positions across images with common backgrounds, allowing for automatic adjustment of lighting parameters and optimal placement of objects using augmented reality overlays, enabling seamless merging and transformation of images with digital filters and beautification techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing image merging techniques are used, then users can combine images together, but the process is time-consuming and requires multiple manual operations
Solution Approach 1:
The system performs automatic object detection, segmentation, and placement without requiring user intervention. The algorithm independently analyzes images, identifies objects, determines optimal positions, and executes the merging process autonomously, eliminating the need for manual masking and positioning operations.
Solution Approach 2:
The system pre-processes images by automatically detecting objects and preparing segmentation masks before the user initiates the merging operation. This preliminary automated preparation eliminates the need for users to perform time-consuming manual preprocessing steps.
2Reliability
If existing merging techniques are used, then images can be combined, but the result often looks unnatural or unconvincing
Solution Approach 1:
The system replaces manual mechanical operations (drag-and-drop, manual masking, feathering adjustments) with automated computer vision algorithms and machine learning models that perform object detection, segmentation, and realistic composite image generation without user intervention.
Solution Approach 2:
The system automatically adjusts critical parameters such as lighting conditions, shadow placement, color balance, and transparency to ensure the merged object blends naturally with the background. These parameter optimizations are performed algorithmically to achieve photorealistic results.
3Ease of manufacture
If manual masking and feathering operations are performed, then image integration can be achieved, but the process becomes complex and inefficient
Solution Approach 1:
The system combines multiple complex operations (object detection, segmentation, masking, feathering, positioning, and blending) into a single automated workflow. This consolidation reduces the number of user-facing steps from multiple manual operations to a single initiation action.
Solution Approach 2:
The system extracts and isolates the object of interest from the source image using automated segmentation algorithms, separating it from the background without requiring user-defined selection tools or manual tracing operations.
Data Source
AI summary
Disclosed are methods and systems for merging of one or more objects from various images into a single image. An example method of combining images comprises obtaining a reference image and obtaining at least one further image. The reference image and the further image have a substantially common background. The further image also includes at least one object such as an individual. The method also includes analyzing the reference image to determine positions of the object for having the object correctly placed in a combined image. Next, the images can be seamlessly merged into a combined image by placing the object from the further image into the reference image according the determined positions. Thereafter, digital filters and beautification techniques can be applied to the combined image.