JLNTV_IPD/CogVideo
4
0
Fork 0
mirror of https://github.com/THUDM/CogVideo.git synced 2025-09-20 21:10:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'main' of https://github.com/THUDM/CogVideo

Browse Source
This commit is contained in:
zR 2024-08-06 11:33:29 +08:00
commit 6141e567e4
2 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -101,7 +101,7 @@ of the **CogVideoX** open-source model.
This folder contains some tools for model conversion / caption generation, etc. This folder contains some tools for model conversion / caption generation, etc.
+ [convert_weight_sat2hf](tools/convert_weight_sat2hf.py): Convert SAT model weights to Huggingface model weights. + [convert_weight_sat2hf](tools/convert_weight_sat2hf.py): Convert SAT model weights to Huggingface model weights.
+ [caption_demo](tools/caption/README.md): Caption tool, a model that understands videos and outputs them in text. + [caption_demo](tools/caption): Caption tool, a model that understands videos and outputs them in text.
## Project Plan ## Project Plan

8
tools/caption/README_zh.md
View File

@ -1,16 +1,16 @@
# 视频字幕 # 视频Caption
通常，大多数视频数据不带有相应的描述性文本，因此需要将视频数据转换为文本描述，以提供必要的训练数据用于文本到视频模型。 通常，大多数视频数据不带有相应的描述性文本，因此需要将视频数据转换为文本描述，以提供必要的训练数据用于文本到视频模型。
## 通过 CogVLM2-Video 模型生成视频描述 ## 通过 CogVLM2-Video 模型生成视频Caption
🤗 [Hugging Face](https://huggingface.co/THUDM/cogvlm2-video-llama3-chat) | 🤖 [ModelScope](https://modelscope.cn/models/ZhipuAI/cogvlm2-video-llama3-chat) | 📑 [Blog](https://cogvlm2-video.github.io/) ｜ [💬 Online Demo](http://cogvlm2-online.cogviewai.cn:7868/) 🤗 [Hugging Face](https://huggingface.co/THUDM/cogvlm2-video-llama3-chat) | 🤖 [ModelScope](https://modelscope.cn/models/ZhipuAI/cogvlm2-video-llama3-chat) | 📑 [Blog](https://cogvlm2-video.github.io/) ｜ [💬 Online Demo](http://cogvlm2-online.cogviewai.cn:7868/)
CogVLM2-Video 是一个多功能的视频理解模型，具备基于时间戳的问题回答能力。用户可以输入诸如 `请详细描述这个视频` 的提示语给模型，以获得详细的视频字幕： CogVLM2-Video 是一个多功能的视频理解模型，具备基于时间戳的问题回答能力。用户可以输入诸如 `请详细描述这个视频` 的提示语给模型，以获得详细的视频Caption：
<div align="center"> <div align="center">
<a href="https://cogvlm2-video.github.io/"><img width="600px" height="auto" src="./assests/cogvlm2-video-example.png"></a> <a href="https://cogvlm2-video.github.io/"><img width="600px" height="auto" src="./assests/cogvlm2-video-example.png"></a>
</div> </div>
用户可以使用提供的[代码](https://github.com/THUDM/CogVLM2/tree/main/video_demo)加载模型或配置 RESTful API 生成视频字幕。 用户可以使用提供的[代码](https://github.com/THUDM/CogVLM2/tree/main/video_demo)加载模型或配置 RESTful API 来生成视频Caption。