Face attribute editing method based on reinforcement learning
A technology of reinforcement learning and attribute editing, which is applied in the field of face attribute editing based on reinforcement learning, can solve problems such as discontinuous attributes, and achieve the effect of saving computing costs
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0046]Such asfigure 1 As shown, a face attribute editing method based on an intensive learning, including the following steps:
[0047](1) Random initialize the intelligent body and the environment and empty the experience playback;
[0048](2) Get the vector representation of the desired training facial face in the potential space of the pre-trained generator model, and use it as an initial state of the environment;
[0049](3) Incorporate the initial state acquired in step (2) into the actor current network, and the calculation of the actor is calculated to obtain a determined state action value, which is the attribute direction vector to be trained;
[0050](4) Implement the status action value obtained in step (3), enter the next state, record this state is the termination state, and the reward of the execution action is obtained;
[0051](5) Whether the previous state in step (2) (3) (4) is performed, the action, reward, and the next state is to be terminated as a five-way group deposit;
[0052...
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