Annotating images
A technology for annotating images and images, applied in still image data retrieval, metadata still image retrieval, instruments, etc., can solve problems such as limited applicability
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0032] Automatically assigning keywords to images allows retrieval, indexing, organization, and understanding of large image data collections. This specification describes techniques for image annotation that treat annotation as a retrieval problem. The described technique utilizes a simple combination of low-level image features and basic distance measures to find the nearest neighbors for a given image. Keywords are then assigned using a greedy label transfer mechanism.
[0033] Image annotation is a difficult task for two main reasons: First, there is a pixel-to-predicate or semantic gap problem where only low-level image features such as color and texture are used to extract Semantically meaningful entities are difficult. Reliably and unambiguously identifying thousands of objects or classes is currently an unsolved problem. The second difficulty arises from the lack of correspondence between image regions and keywords in the training data. For each image, the keywords...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 