Light Field Display Using Device Pose for Dynamic Refocusing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image capture devices lose directional light information, limiting the ability to recover depth and provide personalized views of light field data, which restricts user interaction and experience in visualizing light field images.

Innovation Solution

A method and apparatus that utilize user device pose to dynamically adjust viewpoint and refocusing parameters, allowing natural user interactions to control 4D to 2D re-projection and refocusing of light field images, using sensing devices like IMUs and cameras to estimate device position and orientation, and adapt the focal stack sampling interval based on scene layout.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If conventional image capture devices project a three-dimensional scene onto a two-dimensional sensor, then the device structure is simple and easy to manufacture, but directional light information is lost and depth cannot be recovered

Engineering Contradiction:
Improvedirectional light informationVSAvoiddevice structure
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The image sensor is divided into multiple regions, each corresponding to a specific directional range of incoming light. By segmenting the sensor into spatial zones, the device can capture directional light information from different angles simultaneously, resolving the contradiction between information loss and device complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from conventional 2D image capture to 4D light field capture by adding two spatial dimensions (u, v) to the traditional (x, y) sensor coordinates. This dimensional expansion allows the sensor to record not only position but also directional information of light rays, enabling depth recovery while maintaining a relatively simple device structure

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

2Loss of information

If light-field capture devices measure a four-dimensional light-field by capturing light from different viewpoints, then directional light information is preserved, but the device complexity increases significantly

Engineering Contradiction:
Improvedirectional light informationVSAvoiddevice structure
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

A single image sensor performs multiple functions: it captures both conventional 2D images and 4D light field data by utilizing different regions of the sensor. This multi-functionality eliminates the need for separate devices for different imaging modes, reducing overall system complexity while preserving directional light information

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

Solution Approach 2:

The sensor array automatically captures light from multiple directions simultaneously without requiring mechanical movement or complex optical components. The system uses the inherent spatial arrangement of sensor elements to perform light field measurement, making the device self-sufficient and reducing external complexity

Inventive Principle:
Principle #25Self-service

3Ease of operation

If user interaction is implemented through mouse pointing or touching a tactile screen, then the interface is simple to implement, but the user experience is limited and interaction is not intuitive

Engineering Contradiction:
Improveuser interactionVSAvoiduser experience
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system provides real-time feedback by dynamically adjusting the displayed image's focus and viewpoint based on the user's device orientation and position. As users move or rotate their devices, the light field rendering adapts immediately, creating an intuitive and responsive interaction experience that enhances both ease of operation and adaptability

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent replaces manual mechanical interaction (mouse pointing, touching) with automatic sensor-based tracking of device pose. Inertial sensors and cameras detect user intent through device movement, substituting mechanical control with automated sensing and processing, thereby improving user experience while maintaining operational simplicity

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

4Adaptability or versatility

If multiple users view light field content with different viewpoints, then personalized viewing is achieved, but additional storage is required for multiple focal stacks

Engineering Contradiction:
Improvepersonalized viewingVSAvoidstorage space
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

Instead of storing multiple static focal stacks for different users, the system dynamically generates personalized viewpoints in real-time based on each user's device pose. The light field data is processed on-the-fly to render customized images, eliminating the need for additional storage while maintaining adaptability for multiple users

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes rendering parameters (viewpoint, focus distance, depth of field) dynamically based on user input rather than storing multiple complete image sets. By adjusting these parameters in real-time from a single light field capture, the system achieves personalized viewing without increasing storage requirements

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3099056B1Method and apparatus for displaying a light field based image on a user's device, and corresponding computer program product
Publication Date: 2021.09.08 INTERDIGITAL CE PATENT HOLDINGS SAS
  • EP3099056B1 patent drawingFigure 1~3
  • EP3099056B1 patent drawingFigure 2(a)~2(b)
  • EP3099056B1 patent drawingFigure 4~6

AI summary

A method for displaying at least one light field based image (51) on a user's device (52) is disclosed. Such a method comprises displaying the image focused according to at least one focusing parameter determined as a function of a pose of the user's device (52). A sensor embedded in the users' device, which may be a handheld device, may estimate the pose.