Skip to content

2xHFA2kCompact
Compare
Choose a tag to compare
@Phhofm Phhofm released this 25 Apr 08:16
· 6 commits to main since this release
2xHFA2kCompact
c664eb5
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

Name: 2xHFA2kCompact
Author: Philip Hofmann
Release Date: 18.04.2023
License: CC BY 4.0
Network: SRVGGNetCompact, with parameters: 600,652
Scale: 2
Purpose: A compact anime 2x upscaling model based on musl's HFA2k dataset

Iterations: 93,000
batch_size: 12
HR_size: 384
Epoch: 207 (require iter number per epoch: 214)
Dataset: HFA2k
Number of train images: 2568
OTF Training: Yes
Pretrained_Model_G: 4x_Compact_Pretrain
Training time: 24h+

Description: Compact 2x anime upscaler with otf compression and blur. The '2xHFA2kCompact.pth' (4.6 MB) is the original trained model file, the other model files are conversions using chaiNNer. Trained on musl's latest dataset release for Anime SISR, which has been extracted from modern anime films, where the selection criteria was high SNR, no DOF and high frequency information.

Examples: https://imgsli.com/MTcxNjA4

Example1
Example2
Example3
Example4
Example5
Example6
Example7
Example8
Example9
Example10
Example11
Example12
Example13
Example14
Example15
Example16
Example17
Example18
Example19
Example20
Example21
Example22
Example23
Example24
Example25

Assets 10