-
chatbot
-
face
- deepface: vgg-face
- face_recognition
- inswapper
- refacer
-
super resolution
- CodeFormer: face
-
audio
-
ocr
- paddleOCR
- tesseractOCR
- pororoOCR
- chineseOCR: detection(ctpn) + recognition(DenseNet + CTC)
- cnOCR: PyTorch/MXNet
- HyperLPR: Keras
-
vehicle
- ssd
-
video
- moviepy
- videogrep
- vosk - video transcribe
- backgroundremover: no library supported, just command-line tool
- rembg
-
video + speech recognition
-
Translator
- Korean-English-translator
- Chatgpt-based Koalpaca-Translation-KR2EN
- Chatgpt-based zh2en
- Chatgpt-based ko2en
-
Text2Image
-
Image2Image
-
Speech2Text
-
Text2Speech
- overview: https://platform.openai.com/docs/models/overview
- whisper - speech2text: https://github.com/openai/whisper.git
- https://huggingface.co/models?pipeline_tag=translation&sort=downloads
- https://huggingface.co/models?sort=downloads
- https://github.com/christianversloot/machine-learning-articles
- https://huggingface.co/docs/transformers/pipeline_tutorial
name | base | content |
---|---|---|
llama2-webui | py3 | llama2 - chatbot & code completion |
speechbrain | py3 | text2speech, speech2text |
refacer | py3 | refacer-face swap, inswapper-face swap, codeformer - face restoration |
ocr | py3 | facerecognition - rec/det, cnocr - chn, paddleocr - all, pororoocr - ko/en |
ssd | py2 | chinese_ocr, hyplr-1(chinese license plat recognition), surveillance, vehicle insurance |
nvm | py3 | base image |
opencv3 | py2 | base image |
py3 | base image | |
py2 | base image | |
mongo | base image |
- tensorflow2
- gradio
pip cache purge
apt clean
make clean