Chinese character font automatic generation method based on skeleton guide transmission network
A transmission network and automatic generation technology, applied in the field of computer vision, can solve problems such as uncontrollable, complex structure and font style of Chinese characters, and large vocabulary of Chinese characters
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0042] exist figure 1 Among them, the method for automatically generating Chinese character fonts based on the skeleton-guided transmission network of the present embodiment consists of the following steps:
[0043] (1) Data preprocessing
[0044] Use the mouse to manually extract the font skeleton strokes to form a skeleton font, adjust the skeleton font to an image of 256×256 pixels, save it in png format, merge the calligrapher font and the extracted skeleton font, and construct a set of font datasets with different fonts , in each dataset, calligrapher fonts are on the left, and skeleton fonts are extracted on the right.
[0045] In this step, the font generation is decomposed into a font skeleton synthesis network and a font style correction network to generate a style target font with good stroke detail structure information and a more realistic style.
[0046] (2) Construct font skeleton synthesis network
[0047] The font data set is used as the input of the font sk...
Embodiment 2
[0068] The method for automatically generating Chinese character fonts based on the skeleton-guided transmission network of the present embodiment consists of the following steps:
[0069] (1) Data preprocessing
[0070] This step is the same as in Example 1.
[0071] (2) Construct font skeleton synthesis network
[0072] The font data set is used as the input of the font skeleton synthesis network. During the font generation process, encoder 1 and decoder 2 are constructed. The modules of the same size in encoder 1 and decoder 2 are connected end to end to form the font skeleton synthesis network.
[0073] The encoder 1 of this embodiment is composed of 4 down-sampling modules connected end to end, and the output of the down-sampling modules is 2 from large to small. n ×2 n , 0≤n≤3. The structure of the down-sampling module is the same as that in Embodiment 1.
[0074] The decoder 2 of this embodiment is composed of 4 upsampling modules connected end to end, and the 4 up...
Embodiment 3
[0077] The method for automatically generating Chinese character fonts based on the skeleton-guided transmission network of the present embodiment consists of the following steps:
[0078] (1) Data preprocessing
[0079] This step is the same as in Example 1.
[0080] (2) Construct font skeleton synthesis network
[0081] The font data set is used as the input of the font skeleton synthesis network. During the font generation process, encoder 1 and decoder 2 are constructed. The modules of the same size in encoder 1 and decoder 2 are connected end to end to form the font skeleton synthesis network.
[0082] The encoder 1 of the present embodiment is composed of 10 down-sampling modules connected end to end, and the output of the down-sampling modules is 2 from large to small. n ×2 n , 0≤n≤9. The structure of the down-sampling module is the same as that in Embodiment 1.
[0083] The decoder 2 of this embodiment consists of 10 upsampling modules connected end to end, and th...
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