GPU Point Rendering Gap Filling via Multi-Pass Blending

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing point-based rendering techniques face challenges in reconstructing continuous surfaces without holes as the viewpoint changes, often resulting in impractical solutions or poor image quality.

Innovation Solution

The method involves performing rasterization, filling, and blending passes on an image space using a graphics processing unit (GPU) to fill spaces between pixels with color, creating a contiguous surface without gaps, with geometry parameters updated and blended using a GPU fragment shader, and employing grid spacing that halves after each filling pass.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If point-based rendering is used to represent 3D surfaces, then data efficiency and rendering speed are improved, but gaps and holes appear in the rendered surfaces when viewpoint changes

Engineering Contradiction:
Improverendering speedVSAvoidsurface continuity
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The rendering process is divided into multiple sequential passes (rasterization pass, filling pass, blending pass) that operate in stages. Each pass handles a specific aspect of the rendering task, allowing the system to maintain surface continuity while preserving rendering efficiency through specialized processing at each stage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The filling pass performs preliminary action by identifying and filling gaps in the geometry buffer before the final blending and shading passes. This advance preparation ensures that when the viewpoint changes, the surface continuity has already been established, preventing gaps from appearing in the final rendered image.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If gap filling is performed to create continuous surfaces, then surface continuity is improved, but rendering performance and processing time deteriorate

Engineering Contradiction:
Improvesurface continuityVSAvoidrendering performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The filling pass applies local quality by selectively processing only the regions where gaps exist in the geometry buffer, rather than uniformly processing the entire image. The fragment shader identifies gap regions and applies filling operations only where necessary, maintaining surface continuity while minimizing unnecessary processing overhead.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system performs partial action by implementing a dedicated filling pass that addresses only the specific problem of gap filling, rather than attempting to solve all rendering issues in a single comprehensive pass. This targeted approach improves rendering performance by focusing computational resources on the specific task of maintaining surface continuity.

Inventive Principle:
Principle #16Partial or excessive action

3Manufacturing precision

If multiple rasterization passes are performed to handle different camera views, then view-dependent rendering quality is improved, but processing complexity and time increase

Engineering Contradiction:
Improverendering qualityVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The filling pass implements a universal solution that works across multiple camera views and different rendering scenarios. The same filling logic and fragment shader program are used regardless of the camera position or view angle, reducing processing complexity by avoiding the need for view-specific filling algorithms while maintaining high rendering quality.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3437072B1System and method for rendering points without gaps
Publication Date: 2020.10.14 SONY GROUP CORP
  • EP3437072B1 patent drawingFigure 1~2

AI summary

Various systems and methods disclosed herein are directed to rendering point-based graphics on a computing device with the spaces between points filled in with color to produce the appearance of surfaces without gaps or holes. According to one method, one or more rasterization passes are performed on an image space. One or more filling passes are performed on the pixels in the image space in which the spaces between the pixels are filled with color to form a contiguous surface in a new image plane. One or more blending passes are performed on the image space after the filling passes, in which wherein the color of a group of pixels is a blended together. A new image space is rendered from the image space in the image buffer.