Directional Editing Digital Images Depth Map Feature Volume
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing systems fail to accurately edit digital images due to an inability to describe three-dimensional geometry, leading to inaccuracies in light propagation and shadow creation, especially when only a two-dimensional digital image is available without corresponding geometry.
Innovation Solution
The directional editing technique generates features based on a digital image and depth map, using a ratio of depth in the depth map to depth along a three-dimensional direction, forming a feature volume to improve editing accuracy and propagate light effectively through the image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional ray tracing techniques are used, then lighting and shadow accuracy is improved, but the technique becomes unusable when accurate 3D geometry is unavailable
Solution Approach 1:
The patent creates a synthetic 3D geometry representation by copying and processing depth map information to construct virtual 3D space. Instead of requiring actual 3D geometry, the system synthesizes a workable 3D representation from 2D depth maps, enabling ray tracing to function in scenarios where true 3D geometry is unavailable.
Solution Approach 2:
The system transforms the depth map parameter representation to create a synthetic 3D geometry model. By changing how depth information is represented and processed, the system enables conventional ray tracing techniques to operate on images that originally lacked accurate 3D geometry, thus improving both accuracy and adaptability.
2Adaptability or versatility
If depth maps are used to define 3D geometry, then the technique becomes usable with 2D images, but inaccuracies in depth maps prevent accurate light propagation and shadow creation
Solution Approach 1:
The system performs preliminary processing of the depth map to correct and enhance depth information before using it for ray tracing. By pre-processing the depth data to improve its accuracy and consistency, the system ensures that subsequent light propagation and shadow creation operations achieve higher precision despite starting with imperfect 2D depth map data.
Solution Approach 2:
The patent introduces a synthetic 3D geometry representation as an intermediary between the 2D depth map and the ray tracing process. This intermediary structure mediates the transformation, allowing inaccurate depth map data to be converted into a more reliable 3D representation that enables accurate light propagation and shadow creation.
3Manufacturing precision
If accurate 3D geometry is available, then lighting and shadow edits are accurate, but such geometry is limited in availability for typical real-world images
Solution Approach 1:
The system creates a synthetic copy of 3D geometry from 2D depth map data, enabling the benefits of accurate 3D-based editing to be applied to images that originally lacked such geometry. This copying approach makes high-precision editing accessible across a broader range of images.
Solution Approach 2:
The patent performs a dimensionality transformation by converting 2D depth map information into a synthetic 3D geometry representation. This dimensional change enables the system to apply 3D-based lighting and shadow editing techniques to images that were originally captured in 2D without corresponding geometry.
Data Source
AI summary
Directional propagation editing techniques are described, in one example, a digital image, a depth map, and a direction are obtained by an image editing system. The image editing system then generates features. To do so, the image editing system generates features from the digital image and the depth map for each pixel based on the direction, e.g., until an edge of the digital image is reached. In an implementation, instead of storing a value of the depth directly, a ratio is stored based on a depth in the depth map and a depth of a point along the direction. The image editing system then forms a feature volume using the features, e.g., as three dimensionally stacked features. The feature volume is employed by the image editing system as part of editing the digital image to form an edited digital image.