Cross-domain video semantic concept detection method based on active learning
An active learning and detection method technology, applied in the field of cross-domain video semantic concept detection, can solve the problem of different feature space distribution, and achieve the effect of reducing sample complexity, time complexity, and high classification accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0032] The present invention proposes a cross-domain video semantic concept detection method based on active learning based on active learning. It uses Gaussian random field as a reference classifier, and uses labeled original domain data and unlabeled target domain data as training data. The most uncertain principle in the active learning query strategy selects samples for labeling, adds their new labels to the Gaussian random field, updates the model, and then reselects the new most uncertain samples for labeling, which specifically includes the following steps:
[0033] in figure 1 In the flow chart of the cross-domain active learning algorithm of the present invention, the present invention is mainly divided into four steps: using Gaussian random field as the reference classifier, selecting samples for labeling and updating the reference classifier according to the principle of most uncertainty.
[0034] The first step: using Gaussian random field as the benchmark classifier
[...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap