Probabilistic Image Selection via Pixel Probability Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image editing techniques fail to accurately select image portions due to inaccuracies in user scribbled inputs, especially on smaller screens like touch-enabled tablets and mobile devices, where strokes often cross into unintended areas, leading to incorrect selection.
Innovation Solution
A probabilistic method determines the likelihood of each pixel being part of the selected image portion based on its distance from the center of the scribbled input and differences in visual characteristics, using techniques like graph cuts or Markov Random Fields to segment the intended selection from the rest of the image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional techniques consider user strokes as completely accurate, then the selection process is simple, but the selection accuracy deteriorates when users make imprecise strokes on small screens
Solution Approach 1:
The patent introduces a probabilistic model as an intermediary between the user's scribbled input and the final selection. Instead of directly using the strokes as the selection boundary, the system computes probability maps that represent the likelihood of each pixel being part of the intended selection. This probabilistic intermediary allows the system to handle imprecise user input while still achieving accurate selections by considering multiple possible interpretations of the user's intent.
2Ease of operation
If users make scribbled selections on small touch-enabled screens, then the ease of operation improves, but the selection accuracy deteriorates due to difficulty in keeping strokes within object bounds
Solution Approach 1:
The patent performs preliminary computation of probability maps and segmentation before finalizing the selection. By pre-computing the likelihood of each pixel belonging to the selected portion based on the scribbled input and image characteristics, the system prepares multiple potential selection outcomes in advance. This preliminary action allows the system to anticipate and correct for potential inaccuracies in the user's strokes before committing to a final selection boundary.
3Ease of operation
If the user's finger blocks the view of the object being selected, then the ease of operation is maintained, but the selection accuracy deteriorates
Solution Approach 1:
The patent incorporates feedback mechanisms where the system continuously refines the selection based on the probabilistic model and image characteristics. Even though the user's finger blocks the view, the system provides feedback by computing and using probability maps that reflect the likely intended selection area. The segmentation process uses this probabilistic feedback to adjust the selection boundary, ensuring accuracy despite the lack of direct visual feedback to the user during the selection process.
Data Source
AI summary
Probabilistic determination of selected image portions is described. In one or more implementations, a selection input is received for selecting a portion of an image. For pixels of the image that correspond to the selection input, probabilities are determined that the pixels are intended to be included as part of a selected portion of the image. In particular, the probability that a given pixel is intended to be included as part of the selected portion of the image is determined as a function of position relative to center pixels of the selection input as well as a difference in one or more visual characteristics with the center pixels. The determined probabilities can then be used to segment the selected portion of the image from a remainder of the image. Based on the segmentation of the selected portion from the remainder of the image, selected portion data can be generated that defines the selected portion of the image.