Method for generating image based on multi-target sketch of progressive adversarial generation network
A technology for generating images and multi-objectives, which is applied in the directions of image conversion, image analysis, image enhancement, etc., can solve problems such as errors in synthesis results, and achieve the effect of promoting the overall image and promoting the generation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment
[0076] In this embodiment, XShell, Xftp terminal emulator and Jupyter notebook interactive interface are used on the server of GTX Titan X and 12G graphics card, and the relatively stable deep learning framework tensorflow-GPU version 1.8.0 is used, and the cuda version is 9.0 .176. Using the progressive confrontation generation network proposed by the present invention, combined with the sketch image pair obtained by the sketch discriminative amplification technique, the entire process of generating images from multi-object sketches is completed.
[0077] In this embodiment, 21 categories of MS-COCO image data (including background) are collected. These image categories overlap with the image categories in the Pascal VOC dataset. In order to unify the input of the network, these 20 categories (not included) will be obtained. The image containing the background) is cropped to obtain a size of size=256*256. The instance images of different categories are obtained through the a...
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, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com