17936097. MODULARIZED ATTENTIVE GRAPH NETWORKS FOR FINE-GRAINED REFERRING EXPRESSION COMPREHENSION simplified abstract (International Business Machines Corporation)
Contents
- 1 MODULARIZED ATTENTIVE GRAPH NETWORKS FOR FINE-GRAINED REFERRING EXPRESSION COMPREHENSION
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 MODULARIZED ATTENTIVE GRAPH NETWORKS FOR FINE-GRAINED REFERRING EXPRESSION COMPREHENSION - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
MODULARIZED ATTENTIVE GRAPH NETWORKS FOR FINE-GRAINED REFERRING EXPRESSION COMPREHENSION
Organization Name
International Business Machines Corporation
Inventor(s)
Zhenfang Chen of Cambridge MA (US)
Chuang Gan of Cambridge MA (US)
Dakuo Wang of Cambridge MA (US)
MODULARIZED ATTENTIVE GRAPH NETWORKS FOR FINE-GRAINED REFERRING EXPRESSION COMPREHENSION - A simplified explanation of the abstract
This abstract first appeared for US patent application 17936097 titled 'MODULARIZED ATTENTIVE GRAPH NETWORKS FOR FINE-GRAINED REFERRING EXPRESSION COMPREHENSION
Simplified Explanation
The computer-implemented method described in the abstract involves fine-grained referring expression comprehension using textual expressions and images. Here is a simplified explanation of the patent application:
- Decomposing textual expressions into different modules
- Extracting visual regional proposals from images
- Mining fine-grained object relations using language-guided graph neural networks
- Aggregating matching similarities between textual modules and object relations
Potential Applications
This technology could be applied in various fields such as computer vision, natural language processing, and artificial intelligence.
Problems Solved
This technology helps in understanding and interpreting referring expressions in a more detailed and accurate manner, improving communication between humans and machines.
Benefits
The method enhances the comprehension of complex referring expressions, leading to better object recognition and understanding in visual and textual data.
Potential Commercial Applications
- Enhanced image and text search engines
- Improved virtual assistants for better understanding user queries
Possible Prior Art
There may be prior art related to object recognition, natural language processing, and image-text understanding technologies that could be relevant to this patent application.
Unanswered Questions
How does this technology compare to existing methods in fine-grained referring expression comprehension?
This article does not provide a direct comparison with existing methods in the field.
What are the potential limitations or challenges of implementing this technology in real-world applications?
The article does not address the potential limitations or challenges that may arise when implementing this technology in practical settings.
Original Abstract Submitted
A computer-implemented method for fine-grained referring expression comprehension is provided. The computer-implemented method includes receiving, at a processor, a textual expression and an image as inputs and executing, at the processor, fine-grained referring expression comprehension. The executing includes decomposing the textual expression into different textual modules, extracting visual regional proposals from the image, using language-guided graph neural networks to mine fine-grained object relations from the visual regional proposals and aggregating different matching similarities between the different textual modules and the fine-grained object relations.