Stereo Vision Occlusion via Multi-Layer Depth Shifting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional stereo vision techniques suffer from occlusion issues when constructing scenes from different viewpoints, resulting in a lack of realism due to overlapping objects that do not change appropriately when the view is shifted.

Innovation Solution

The method involves segmenting images into layers and shifting them at different speeds, using depth images and object masks to interpolate views between positions, with the object layer shifting faster than the background layer, effectively addressing occlusion and enhancing realism.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional panoramic techniques are used to construct stereo vision views, then the scene can be constructed from multiple viewpoints, but occlusion occurs during view movement resulting in a lack of realism

Engineering Contradiction:
Improverealism of stereo visionVSAvoidocclusion effect
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The image is divided into multiple layers based on depth information from depth images. Objects are segmented into separate layers from the background using thresholding techniques applied to depth data. This segmentation allows different layers to be shifted independently, resolving occlusion issues by properly ordering layers according to their depth positions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces depth as an additional dimension by utilizing depth images to create multiple layers. Instead of treating the image as a single 2D plane, the solution creates a layered structure with depth information, allowing objects at different depths to be processed and shifted independently, thus eliminating occlusion artifacts.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If images are cut into several layers and shifted at different speeds, then occlusion is resolved and realism is improved, but computational complexity increases

Engineering Contradiction:
Improverealism of stereo visionVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The image is divided into multiple layers based on depth information from depth images. Objects are segmented into separate layers from the background using thresholding techniques applied to depth data. This segmentation allows different layers to be shifted independently, resolving occlusion issues by properly ordering layers according to their depth positions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces depth as an additional dimension by utilizing depth images to create multiple layers. Instead of treating the image as a single 2D plane, the solution creates a layered structure with depth information, allowing objects at different depths to be processed and shifted independently, thus eliminating occlusion artifacts.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10204418B2Method and apparatus for facilitating stereo vision through the use of multi-layer shifting
Publication Date: 2019.02.12 NOKIA TECHNOLOGIES OY
  • US10204418B2 patent drawing
  • US10204418B2 patent drawing
  • US10204418B2 patent drawing

AI summary

A method, apparatus and computer program products are provided for facilitating stereo vision through the use of multi-layer shifting. One example method includes receiving a first RGB image and a second RGB image, each captured from a different position, generating corresponding depth images, performing a first segmentation on the depth images, performing a second segmentation on the RGB images resulting in a first object mask and a second object mask, stitching of the background layers to form a background layer, generating a series of images comprising a first image representative of a view from the first position and a second image representative of a view from the second position, the series of images demonstrating a shifting of an object layer occurring at a higher speed than a shifting of the background layer.