Occluded Object Feature Estimation via Temporal Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Static models are inadequate for accurately representing dynamic environments, where objects and users interact, leading to incomplete modeling of occluded objects and inaccurate interaction identification.
Innovation Solution
The system estimates missing features of occluded objects in a dynamic environment using data from sensors, such as cameras, and a catalog of reference objects, by comparing occluded object data with previously captured images and reference object data to generate a more accurate model of the environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If static modeling is used, then device complexity is reduced, but measurement precision of dynamic environments deteriorates
Solution Approach 1:
The patent transforms static modeling into dynamic modeling by continuously updating object representations over time. The system tracks objects through multiple frames, maintaining temporal consistency while adapting to motion. This resolves the contradiction by introducing dynamic processing that improves accuracy without requiring entirely new computational frameworks.
Solution Approach 2:
The patent performs preliminary actions by capturing and storing object features at multiple time points before occlusion occurs. Historical data about objects is accumulated and used to infer occluded portions, allowing the system to prepare reference information in advance that aids subsequent occlusion handling.
2Loss of time
If objects are modeled as static, then processing time is reduced, but loss of information about occluded objects increases
Solution Approach 1:
The system captures and stores object features at multiple time points before occlusion occurs. Historical data about objects is accumulated and used to infer occluded portions, allowing the system to prepare reference information in advance that aids subsequent occlusion handling.
Solution Approach 2:
The patent uses feedback from temporal data by comparing current frame information with historical data from previous frames. This feedback mechanism allows the system to identify occluded regions by detecting changes in object appearance over time and use historical information to fill in missing features.
3Measurement precision
If dynamic modeling is implemented, then measurement precision of occluded objects improves, but device complexity increases
Solution Approach 1:
The patent transforms static modeling into dynamic modeling by continuously updating object representations over time. The system tracks objects through multiple frames, maintaining temporal consistency while adapting to motion. This resolves the contradiction by introducing dynamic processing that improves accuracy without requiring entirely new computational frameworks.
Solution Approach 2:
The system creates a temporal copy of object data from previous frames to infer current occluded states. By copying and comparing historical object representations with current observations, the system can reconstruct occluded portions without requiring complex real-time sensing for every possible view.
4Measurement precision
If multiple temporal data is processed, then accuracy of environmental model improves, but use of energy increases
Solution Approach 1:
The system processes only the necessary temporal data required for occlusion handling rather than all available data. By identifying which historical frames and object features are relevant to current occlusion situations, the system performs partial processing that achieves sufficient accuracy without excessive energy consumption.
Data Source
AI summary
A number of images of an environment may be obtained over time. In some images, a portion of an object included in the environment may be occluded by another object in the environment. In these images, features of the occluded portion of the object may be absent. The absent features may be estimated based on features generated from previous images including the object where the occluded portion of the object was not occluded in the previous images. In some cases, the absent features may be estimated based on data associated with a reference object that corresponds to the occluded object with the reference object being included in a catalog of reference objects.