Virtual Viewpoint Image Generation for 3D Trajectory Display

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing techniques for generating virtual viewpoint images struggle to accurately display the three-dimensional movement path of objects, as the movement trajectory is typically superimposed on two-dimensional images, making it difficult to acquire and display effectively.

Innovation Solution

An image processing apparatus that acquires positional information about an object, generates trajectory data, and creates a virtual viewpoint image based on this data, using a combination of shape estimation, movement trajectory data addition, and rendering processes to produce a three-dimensional representation of the object's movement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If movement trajectory is superimposed on two-dimensional images, then the trajectory information can be displayed, but it is difficult to accurately represent the three-dimensional movement path

Engineering Contradiction:
Improvethree-dimensional movement path informationVSAvoidtrajectory display accuracy
Core Design Contradiction:
Loss of informationVSManufacturing precision

Solution Approach 1:

The patent transitions from two-dimensional image superposition to three-dimensional virtual space reconstruction. By building a three-dimensional model of the object and representing the movement trajectory in virtual space with multiple coordinates (x, y, z positions across multiple viewpoints), the system accurately captures and displays the three-dimensional movement path that cannot be represented in 2D images alone.

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

2Measurement precision

If multiple imaging apparatuses are arranged to surround the imaging area, then three-dimensional information can be acquired, but the system complexity increases

Engineering Contradiction:
Improvethree-dimensional positional information accuracyVSAvoidimaging system configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines multiple captured images from different viewpoints into a unified three-dimensional model. By merging the positional information from multiple imaging apparatuses into a single virtual viewpoint image generation system, the patent achieves accurate three-dimensional trajectory representation while managing system complexity through integrated processing.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a virtual copy of the physical imaging system by generating virtual viewpoint images that simulate what would be captured from any viewpoint within the imaging area. This virtual modeling approach allows three-dimensional trajectory analysis without requiring physical access to every possible imaging position.

Inventive Principle:
Principle #26Copying

3Manufacturing precision

If virtual viewpoint image generation is used, then three-dimensional representation is achieved, but the processing complexity increases

Engineering Contradiction:
Improvetrajectory representation accuracyVSAvoidimage processing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary processing by pre-establishing the three-dimensional model of the object and pre-calculating the relationship between multiple viewpoints and the object's positional information. This preliminary setup enables efficient trajectory calculation and virtual viewpoint image generation, reducing the processing complexity during actual trajectory analysis.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20240296615A1Image processing apparatus for generating virtual viewpoint image, image processing method, and storage medium
Publication Date: 2024.09.05 CANON KK
  • US20240296615A1 patent drawing
  • US20240296615A1 patent drawing
  • US20240296615A1 patent drawing

AI summary

An image processing apparatus includes one or more memories storing instructions, and one or more processors executing the instructions to acquire data including positional information about an object, generate trajectory data based on the positional information about the object, and generate a virtual viewpoint image based on the trajectory data.