Live View Annotation Capture for AI Training Image Collection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The shortage of image data and annotators for AI training data creation poses a significant challenge, leading to a burden in acquiring new images for annotation.
Innovation Solution
An imaging apparatus that receives annotation information, displays it with a live view image, allows user selection, and generates annotation data by associating selected information with captured images, thereby efficiently collecting images that meet specific annotation requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If new photographs are captured to secure images for annotation, then the shortage of annotation data is addressed, but the workload and burden of acquiring images increases
Solution Approach 1:
The system performs preliminary actions by displaying annotation information and captured images in real-time during the imaging process, allowing annotators to make selections and associations before final data collection. This real-time preliminary processing reduces the need for separate post-processing steps and decreases overall acquisition workload.
Solution Approach 2:
The imaging apparatus performs self-service by automatically capturing images, displaying annotation information, and generating annotation data through user selection without requiring external intervention. The system serves itself by integrating multiple functions (imaging, display, data generation) within a single device, reducing the burden of acquiring images from multiple sources.
2Productivity
If annotation information is added to synthesized images, then training data creation is promoted, but the fundamental shortage of image data and annotators remains unsolved
Solution Approach 1:
The imaging apparatus performs multiple functions including capturing images, displaying annotation information, receiving user selections, and generating annotation data. This multi-functionality consolidates what would otherwise require separate systems and personnel, addressing the shortage of image data and annotators by enabling a single device to handle the entire annotation workflow.
Solution Approach 2:
The system implements feedback by displaying annotation information and captured images in real-time, allowing annotators to immediately see and select from available options. This real-time feedback loop enables efficient annotation data generation without requiring large pre-existing datasets or multiple annotators, as the system guides the annotation process interactively.
Data Source
AI summary
An imaging apparatus executes reception processing in which annotation information related to an image is received, executes first display control processing in which a live view image is displayed on a display unit, executes second display control processing in which the annotation information received by the reception processing is associated with the live view image and displayed on the display unit, executes imaging processing in which a captured image is generated, executes input processing in which a selection operation for selecting the annotation information displayed on the display unit is accepted, and executes data generation processing in which annotation data, in which the annotation information selected by the selection operation and the captured image generated by imaging processing are associated, is generated.