https://tesseract-ocr.github.io/tessdoc/Home.html
🥳~ (精简)
sudo dnf install --best --allowerasing tesseract.x86_64 tesseract-langpack-ell.noarch tesseract-langpack-ell.noarch tesseract-langpack-jpn.noarch tesseract-langpack-chi-sim.noarch tesseract-langpack-chi-tra.noarch 🥳~ (精简)
sudo dnf install tesseract.x86_64 tesseract-langpack-ell tesseract-langpack-jpn tesseract-langpack-chi-sim tesseract-langpack-chi-tratessdoc
Tesseract User Manual
Tesseract documentation
[DuangSUSE@duangsuse]~/Projects/extract-subtitles% you-get 'https://www.youtube.com/watch?v=ex3HRrxEcLE'
site: YouTube
title: [Synth V] Something New (Aku P feat. Eleanor Forte)
stream:
- itag: 137
container: mp4
quality: 1920x1080 (1080p)
size: 8.6 MiB (9007866 bytes)
# download-with: you-get --itag=137 [URL]
YouTube
[Synth V] Something New (Aku P feat. Eleanor Forte)
Demo project for Synthesizer V.
https://synthesizerv.com
https://synthesizerv.com
extract_subtitles.py
7.6 KB
[DuangSUSE@duangsuse]~/Projects/extract-subtitles% ./extract_subtitles.py -crop '(555,912)(726,105)' --crop-debug \[Synth\ V\]\ Something\ New\ \(Aku\ P\ feat.\ Eleanor\ Forte\).mp4
[Video Path] [Synth V] Something New (Aku P feat. Eleanor Forte).mp4
[Frame Directory] frames
[Subtitle Language] chi_sim
Extracting key frames...
429 13
How can something be so nice?
How can Something be so nice?
今天来给大家分享个脚本…… 是我从 https://github.com/shawnsky/extract-subtitles 重构的版本
原版几乎不能用(滑稽
经由本苏久坐劳累后妙手回春就又可以用了(跑
原版几乎不能用(滑稽
经由本苏久坐劳累后妙手回春就又可以用了(跑
GitHub
GitHub - shawnsky/extract-subtitles: Extract Subtitles From Video 视频字幕提取 帧间差分法识别关键帧 OCR识别
Extract Subtitles From Video 视频字幕提取 帧间差分法识别关键帧 OCR识别 - GitHub - shawnsky/extract-subtitles: Extract Subtitles From Video 视频字幕提取 帧间差分法识别关键帧 OCR识别
Forwarded from dnaugsuz
其实这个项目本来就不适合用 Python 写
原作者内存优化得太垃圾,把所有帧都放内存里…… 我这 4G 处理个 4 分钟压缩后 4M 的小视频都撑不过 1:00
我正在考虑是添加一个 gc 内存压力提示呢,还是用其他语言重写呢,还是换 pypy 呢 🤔
反正最后一部分 cv 是不可以继续用 Python 的,内存压力实在是太大了……
原作者内存优化得太垃圾,把所有帧都放内存里…… 我这 4G 处理个 4 分钟压缩后 4M 的小视频都撑不过 1:00
我正在考虑是添加一个 gc 内存压力提示呢,还是用其他语言重写呢,还是换 pypy 呢 🤔
反正最后一部分 cv 是不可以继续用 Python 的,内存压力实在是太大了……
duangsuse::Echo
🐶 Sticker
我觉得可以做一个 GUI,来支持时轴框选歌词区域。
horizontalLayout {
TextureView,
Sidebar(verticalLayout {
playerTime,
horizontalLayout { playerControl, btnNewCrop }
ListView(crops/*time rect*/),
TextView /*current_subtitle*/,
TextArea /*time_subtitles*/
})
}#pragma once
#include <QMainWindow>
namespace Ui {
class MainWindow;
}
class MainWindow: public QMainWindow { Q_OBJECT
public:
MainWindow(QWidget* parent = nullptr);
~MainWindow();
private slots:
void onShowToastClicked();
void onShowProgressSpinnerClicked();
void onShowProgressHorizontalClicked();
private:
Ui::MainWindow* ui;
}; 🤔 很久没写 Qt 了,应该默写下这个GitHub
duangsuse-valid-projects/Qt-Qt5-Widgets-Intro
:thinking:. Contribute to duangsuse-valid-projects/Qt-Qt5-Widgets-Intro development by creating an account on GitHub.
Forwarded from dnaugsuz
https://github.com/lars76/chinese-subtitle-ocr
找到了个更好的,不必再岔出去写什么 GUI 了……
找到了个更好的,不必再岔出去写什么 GUI 了……
GitHub
GitHub - lars76/chinese-subtitle-ocr: Optical character recognition for Chinese subtitles using SSD and CNN
Optical character recognition for Chinese subtitles using SSD and CNN - GitHub - lars76/chinese-subtitle-ocr: Optical character recognition for Chinese subtitles using SSD and CNN
add_argument(name, metavar, nargs, action, type, default, help) 🤔
duangsuse::Echo
add_argument(name, metavar, nargs, action, type, default, help) 🤔
原作者混合了 cv2 和 Python PIL 图像处理,我给它改回纯 cv2,方便使用……
extract_subtitles.py
7.8 KB
我给改回来了,现在用的
cv2.imread 和 cv2.UMat 的 slice#Python #DontKnow https://stackoverflow.com/questions/18946662/why-shouldnt-i-use-pypy-over-cpython-if-pypy-is-6-3-times-faster
Stack Overflow
Why shouldn't I use PyPy over CPython if PyPy is 6.3 times faster?
I've been hearing a lot about the PyPy project. They claim it is 6.3 times faster than the CPython interpreter on their site.
Whenever we talk about dynamic languages like Python, speed is one of t...
Whenever we talk about dynamic languages like Python, speed is one of t...