Motion Vector Optimization Across Refractive and Reflective Interfaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing graphics engines struggle to generate accurate motion vectors for translucent objects in path-traced images, particularly when multiple refractive interfaces are involved, leading to noise and artifacts due to complex light reflections and refractions.
Innovation Solution
The system determines accurate motion vectors by treating multiple refractive interfaces as a single morphing layer, using numerical optimization methods like Newton's quadratic method to minimize the angle between foreground and background positions, and applies these vectors for image denoising and temporal anti-aliasing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If path tracing is used to achieve physically-based rendering with long specular chains and multiple interfaces, then rendering quality and realism are improved, but computing resource consumption increases significantly
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and storing motion vectors for refractive interfaces during the rendering process. These pre-computed motion vectors are then reused in subsequent frames, eliminating the need for repeated complex calculations and reducing overall computing resource consumption while maintaining high rendering quality.
Solution Approach 2:
The patent creates copies of motion vector data from refractive interfaces and reuses them across multiple frames. By storing and reutilizing these motion vectors, the system avoids redundant calculations, significantly reducing computing resource requirements while preserving the physically-based rendering quality achieved through path tracing.
2Measurement precision
If multiple refractive interfaces are processed individually with repeated calculations, then accurate motion vectors can be generated, but processing time and computational complexity increase
Solution Approach 1:
The system performs preliminary computation of motion vectors for refractive interfaces and stores them for later use. This pre-computation approach ensures accurate motion vectors are available when needed without requiring repeated complex calculations, thereby reducing processing time while maintaining measurement precision.
Solution Approach 2:
The patent copies and reuses motion vector data from refractive interfaces across multiple frames instead of recalculating them each time. This copying strategy maintains the accuracy of motion vectors while significantly reducing the time required for processing, as the same accurate data is reused rather than recomputed.
3Use of energy by moving object
If insufficient rays per pixel are used in path tracing, then computing resources are saved, but image quality deteriorates with spotty results and noise
Solution Approach 1:
The patent uses copied motion vector data from previous frames to guide ray tracing in current frames. This allows the system to achieve better image quality with fewer rays per pixel, as the motion vector information provides additional guidance that compensates for the reduced number of samples, thereby lowering computing resource requirements while maintaining image quality.
Data Source
AI summary
Systems and methods relate to the determination of accurate motion vectors, for rendering situations such as a noisy Monte Carlo integration where image object surfaces are at least partially translucent. To optimize the search for “real world” positions, this invention defines the background as first path vertices visible through multiple layers of refractive interfaces. To find matching world positions, the background is treated as a single layer morphing in a chaotic way, permitting the optimized algorithm to be executed only once. Further improving performance over the prior linear gradient descent, the present techniques can apply a cross function and numerical optimization, such as Newton's quadratic target or other convergence function, to locate pixels via a vector angle minimization. Determined motion vectors can then serve as input for services including image denoising.