Image Capturing Apparatus Learned Model Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image capturing apparatuses face challenges in managing multiple learned models for subject identification, as they require significant storage capacity and efficient switching mechanisms, especially with the expanding types of subjects and applications like autofocusing, auto exposure, and image processing.
Innovation Solution
An image capturing apparatus system that connects with an external device to receive, select, and transmit learned models, utilizing a reception unit, selection unit, and transmission unit to manage and switch between multiple learned models based on list information and user requests, optimizing storage and processing capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple learned models are held on the image capturing apparatus to support various subjects and processing functions, then the subject identification capability and processing versatility are improved, but the recording capacity requirement increases significantly
Solution Approach 1:
The learned models are segmented between external storage and the image capturing apparatus. The apparatus maintains only essential model information locally while storing complete models externally, dividing the storage burden into manageable segments that match the apparatus's actual needs.
Solution Approach 2:
An information processing apparatus acts as an intermediary between external storage and the image capturing apparatus. This mediator manages the transmission and selection of learned models, allowing the camera to access multiple models without permanently storing them all, thus resolving the storage capacity conflict.
2Speed
If all learned models are stored on the image capturing apparatus to ensure immediate availability, then the processing speed and responsiveness are improved, but the device complexity and storage management become difficult
Solution Approach 1:
The system performs preliminary actions by pre-selecting and transmitting only the necessary learned models to the image capturing apparatus based on predicted usage scenarios. This advance preparation ensures fast access speed while minimizing the number of models that need to be managed on-device.
Solution Approach 2:
The image capturing apparatus automatically determines which learned models are needed based on shooting conditions and subject types, then requests only those specific models from external storage. This self-service approach eliminates manual storage management complexity while maintaining fast access to relevant models.
3Adaptability or versatility
If the image capturing apparatus maintains a large library of learned models for future subject types, then the future-proofing and adaptability are improved, but the current storage requirements and cost increase
Solution Approach 1:
The external storage system serves multiple functions: storing current learned models, archiving future models, and providing on-demand transmission. This universal storage solution supports both current and future subject types without requiring the apparatus to pre-allocate space for all possible future models.
Solution Approach 2:
The learned model library is made dynamic through on-demand transmission. The apparatus can adapt to future subject types by receiving new models when needed, rather than maintaining a static, pre-filled storage capacity. This dynamic approach allows future-proofing without current storage overhead.
Data Source
AI summary
An image capturing apparatus includes a reception unit configured to connect with an external device, which is able to transmit a plurality of learned models, and receive list information of a plurality of learned models, a selection unit configured to select, based on the list information of the plurality of learned models, a learned model from the plurality of learned models, and a transmission unit configured to transmit a transmission request for the learned model selected by the selection unit to the external device, wherein, the reception unit receives the selected learned model transmitted from the external device.