Ask-Anything icon indicating copy to clipboard operation
Ask-Anything copied to clipboard

[CVPR2024 Highlight][VideoChatGPT] ChatGPT with video understanding! And many more supported LMs such as miniGPT4, StableLM, and MOSS.

Results 78 Ask-Anything issues
Sort by recently updated
recently updated
newest added

There is also another work called video ChatCaptioner. It looks that these two ideas are very related. Can you tell the main difference between your work and Video ChatCaptioner? https://github.com/Vision-CAIR/ChatCaptioner/tree/main/Video_ChatCaptioner

你知道Tekla open API吗

主要报错是Internal: src/sentencepiece_processor.cc(1101) [model_proto->ParseFromArray(serialized.data(), serialized.size())] 报错行: File "/opt/data/private/LH/Ask-Anything/video_miniGPT4/demo_video.py", line 58, in model = model_cls.from_config(model_config).to('cuda:1') File "/opt/data/private/LH/Ask-Anything/video_miniGPT4/minigpt4/models/mini_gpt4.py", line 243, in from_config model = cls( File "/opt/data/private/LH/Ask-Anything/video_miniGPT4/minigpt4/models/mini_gpt4.py", line 86, in __init__ self.llama_tokenizer =...

Hi, awesome work, how can we support an ability to use real-time video (streamed video)?

您好,请问image_reasoning - clevr数据集具体是哪个?我按文章中的引用找到了https://cs.stanford.edu/people/jcjohns/clevr/,下载了[CLEVR v1.0 (18 GB)],解压后发现图片内容和json中的格式不对应。

![image](https://github.com/OpenGVLab/Ask-Anything/assets/144189381/6c413f08-54c5-4f8f-9b6a-93e6691f6466) When I train the stage1 with command `bash scripts/run_7b_stage1.sh`, process returns the above error, is there any method to solve the problems?