Image Sequence Approximation via Pixel Subset Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In digital image production, there is a need for techniques that can approximate a fully calculated image sequence without calculating all image data, as fully calculating all data is often constrained by time or computational resources, especially when only a low-quality preview is desired.
Innovation Solution
A computer-implemented method that involves fully calculating a first image in a sequence and then determining a subset of pixels to calculate in subsequent images, propagating calculated image data from neighboring images to non-calculated pixels, and interpolating image data to approximate the remaining pixels, using render masks and motion data to identify occluded pixels and manage pixel ages for quality preservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If all image data is fully calculated, then image quality is maintained, but time and computational resources are excessively consumed
Solution Approach 1:
The patent segments the image rendering process into two distinct parts: (1) fully calculating only the first image in the sequence, and (2) approximating subsequent images by propagating data from already-calculated images. This segmentation allows the system to maintain high quality for the base image while significantly reducing computational effort for the remaining images in the sequence.
Solution Approach 2:
The patent applies partial action by calculating only the necessary portion of image data (the first image fully, and selectively propagating to subsequent images) rather than fully calculating every image in the sequence. This partial calculation approach maintains acceptable image quality while dramatically reducing the total computational burden and rendering time.
2Measurement precision
If all image data is fully calculated, then complete image sequence accuracy is achieved, but computational resources are excessively consumed
Solution Approach 1:
The patent divides the computational workload by segmenting which images receive full calculation versus approximation. The first image undergoes complete calculation to establish accurate baseline data, while subsequent images use propagation from this base, significantly reducing total computational resource consumption while maintaining sufficient accuracy for the complete sequence.
Solution Approach 2:
The system performs partial calculation by fully processing only the first image and using data propagation for the remainder of the sequence. This partial action approach achieves acceptable accuracy for the complete image sequence while consuming far fewer computational resources than full calculation of every image.
3Productivity
If a subset of pixels is calculated in subsequent images, then rendering speed increases, but image quality may degrade
Solution Approach 1:
The patent performs preliminary action by fully calculating the first image in the sequence before generating subsequent images. This pre-calculated base image provides accurate reference data that is then propagated to subsequent frames, ensuring that even though not all pixels in later images are fully calculated, the rendered quality remains acceptable through the use of this preliminary, high-quality reference data.
Data Source
AI summary
The disclosure provides an approach for approximating a fully calculated image sequence from a subset of calculated image data. In one configuration, a rendering application calculates, for each image in the image sequence, a subset of pixels including pixels previously occluded by object(s) in preceding images and pixels whose ages have exceeded a threshold value. The rendering application then approximates the remaining non-calculated image data by interpolating the image data that has been calculated and merges the calculated and approximated image data to obtain the final image sequence. In a particular configuration, entire images may be recalculated at regular intervals to prevent excessive image quality degradation. In another configuration, the recalculation of pixel data may be offset by randomly initializing pixel age values. In yet another configuration, a combination of recalculating entire images and refreshing pixels at offset intervals may be used.