Options -> Preferences
Project -> Track/Send Defaults
Modify the Record config
Cancel the "Monitor Input"
yuv420p
YUV420P又叫plane平面模式,Y , U , V分别在不同平面,也就是有三个平面,它是YUV标准格式4:2:0,主要分为:YU12和YV12
YU12
在android平台下也叫作I420格式,首先是所有Y值,然后是所有U值,最后是所有V值
YV12
YV12格式与YU12基本相同,首先是所有Y值,然后是所有V值,最后是所有U值。只要注意从适当的位置提取U和V值,YU12和YV12都可以使用相同的算法进行处理。
NV12
- 安装依赖
# yum install autoconf automake bzip2 bzip2-devel cmake freetype-devel gcc gcc-c++ git libtool make pkgconfig zlib-devel libvorbis-devel wget
- 建立需要用来存放源码、build和可执行文件的文件夹
# 存放各种组件的源码包
mkdir -p ~/ffmpeg_sources
# 存放编译后的文件
mkdir -p ~/ffmpeg_build
- Uninstall cuda
run cuda-uninstaller in /usr/local/cuda/bin using root user
- Install Cuda
Download runfile from cuda official website.
sudo sh cuda_xxx.run.
Note that don't select nvidia driver! - Configuration add these lines to the end of ~/.bashrc
self = <torch.utils.file_baton.FileBaton object at 0x7f7418fe1fa0>
def wait(self):
'''
Periodically sleeps for a certain amount until the baton is released.
The amount of time slept depends on the ``wait_seconds`` parameter
passed to the constructor.
'''
RPC failed; curl 18 transfer closed with outstanding read data remaining.
SOLUTION:
(venv) sdtv@sdtv-Precision-7920-Tower:~/MGLD-VSR$ git config --list --global
http.postbuffer=1048576000
http.maxrequestbuffer=100M
http.version=HTTP/2
core.compression=0
pip install -e git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers -e git+https://github.com/openai/CLIP.git@main#egg=clip
Manually install
git clone https://github.com/CompVis/taming-transformers.git
git clone https://github.com/openai/CLIP.git
cd /path/to/taming-transformers
pip install -e .
- Download Models
(1) pretrained diffusion denoising U-net and video variational autoencoder
https://connectpolyu-my.sharepoint.com/:f:/g/personal/19046191r_connect_polyu_hk/EvI_j1SUiVFBlwEy4i62ckgB1XEHeqfFcJS4Ho6JQrTAWA?e=rDT4M4
or
https://pan.baidu.com/s/1xQF996RsxnmN-60ZLB6Vig?pwd=gh4i
downloaded files:
mgldvsr_unet.ckpt
video_vae_cfw.ckpt
https://github.com/NVIDIA/MinkowskiEngine/issues/543
export MAX_JOBS=1
- PASD - make photo clearer, personalized styplization, colorization
https://github.com/yangxy/PASD - DA-CLIP - image dehazing(去雾), image deblurring, face inpainting(修复面部图像的缺陷),
image desnowing(除雪), image shadow removal(图像去阴影), JPEG arifact deduction, low-light image enhancement,
image denoising, image raindrop removal(玻璃上的雨滴去除), image deraining(图像中去除下雨)
https://github.com/Algolzw/daclip-uir?tab=readme-ov-file - SeeSR: Real-World Image Super-Resolution
https://github.com/cswry/SeeSR