Object Recognition Feature Highlighting for Interactive Result Display
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object recognition systems suffer from poor user interactivity and user experience, particularly in applications involving image analysis, due to the lack of effective ways to present recognition results beyond simple text descriptions.
Innovation Solution
A method and device for object recognition that determines a target feature part of an object and displays both image and associated feature information, using pre-trained models to enhance user interaction and understanding of recognition results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If simple text descriptions are used to present recognition results, then the system is simple and fast, but user interactivity and user experience are poor
Solution Approach 1:
The patent combines multiple modes of information presentation (text descriptions, images, and video clips) into a unified recognition result display system. The processing module integrates these different media types to provide comprehensive information about recognized objects, thereby improving user interactivity and experience without creating separate independent systems for each media type.
Solution Approach 2:
The patent transitions from one-dimensional text-only descriptions to multi-dimensional information presentation by incorporating visual elements (images and video clips). This dimensional expansion allows users to perceive recognition results through multiple sensory channels, enhancing user experience and interactivity while maintaining system coherence.
2Reliability
If only basic recognition results are displayed, then the system is simple and fast, but user experience and trust are insufficient
Solution Approach 1:
The patent segments the recognition result information into distinct components: text descriptions for identification, images for visual confirmation, and video clips for dynamic observation. This segmentation allows each information type to serve its specific purpose in building user trust while maintaining overall system efficiency. Users can access different levels of detail based on their needs.
Solution Approach 2:
The system performs preliminary processing to select and prepare relevant information (images, video clips) before presenting recognition results. This preliminary action ensures that only meaningful and relevant information is displayed, reducing information overload while maintaining completeness. The system proactively determines which information elements will most effectively build user trust based on the recognized object type and context.
3Ease of operation
If comprehensive information is presented, then user experience and trust are improved, but the system becomes more complex and slower
Solution Approach 1:
The patent implements a dynamic information presentation system where the processing module adapts the amount and type of information displayed based on user interactions and context. The system can provide comprehensive information when needed while maintaining speed by only presenting essential information when appropriate. This dynamic adjustment allows the system to balance between completeness and performance based on real-time conditions.
Solution Approach 2:
The patent applies different levels of information detail to different aspects of recognition results based on local requirements. For example, highly detailed visual information may be provided for certain object types while more summarized text descriptions are used for others. This localized quality adjustment optimizes the balance between user experience and system speed by providing comprehensive information only where it adds the most value.
Data Source
AI summary
Disclosed are a method and a device for object recognition. The method includes: obtaining an object image, wherein the object image includes one or more objects to be recognized; for each object to be recognized, using a pre-trained object recognition model to generate a recognition result of the object to be recognized, and determining a target feature part of the object to be recognized according to the recognition result; and displaying a first screen, wherein the first screen includes at least a part of the object image and target feature information associated with the target feature part.