AR Object Augmentation with Consistent Lighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality devices cannot consistently incorporate user alterations into multiple perspectives of a physical scene, limiting the dynamic and realistic augmentation of virtual objects within a real-world environment.
Innovation Solution
The method involves capturing a visual scene using camera devices, identifying predefined objects, retrieving their visual characteristics, and rendering frames that augment the objects' appearance based on both the captured scene and predefined profiles, including lighting effects and shadows, to create a more realistic and adaptive augmented reality experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If virtual objects are inserted into captured images, then the user can interact with augmented reality content, but the alterations cannot be consistently incorporated into multiple perspectives of the physical scene
Solution Approach 1:
The system continuously captures images from multiple perspectives, tracks the physical object across these images, and uses feedback from object identification and characteristic retrieval to consistently apply virtual alterations across all perspectives. The feedback loop ensures that as the camera moves and captures different angles, the virtual objects remain correctly positioned and adapted to each new perspective while maintaining consistency with the original user alterations.
2Area of stationary object
If the system captures and processes multiple images from different perspectives, then it can provide comprehensive scene coverage, but the computational complexity and processing time increase
Solution Approach 1:
The system extracts only the essential elements from captured images - specifically identifying physical objects and their characteristics - rather than processing entire images. By extracting object identifiers, retrieving pre-stored visual characteristics, and focusing computational resources only on objects that require augmentation, the system reduces processing complexity while maintaining comprehensive scene coverage across multiple perspectives.
3Manufacturing precision
If the system retrieves and applies detailed visual characteristics information, then the augmentation realism improves, but the data retrieval and processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-storing visual characteristics information for various object types in a database during system setup or object registration phases. When processing captured images, the system quickly retrieves pre-computed characteristics rather than calculating them in real-time. This preliminary preparation of object characteristic data significantly reduces processing time during actual augmented reality rendering while maintaining high augmentation realism through detailed visual properties.
Data Source
AI summary
Techniques for augmenting an appearance of a first object. Embodiments include capturing a visual scene for display. Here, the visual scene includes a physical object and wherein the visual scene is captured using one or more camera devices. The physical object is identified as a first predetermined object type, based on one or more object identifiers associated with the physical object. Embodiments also retrieve visual characteristics information corresponding to the first predetermined object type. A sequence of frames that includes the first object is then rendered for display, where the appearance the first object in the rendered sequence of frames is augmented based on the retrieved visual characteristics information and an appearance of the physical object in the captured visual scene.