Request for a file in the repository

patrick-tssn / VSTAR

[ACL2023] VSTAR is a multimodal dialogue dataset with scene and topic transition information

https://vstar-benchmark.github.io/

12 stars 2 forks source link

Request for a file in the repository #1

Open Shrad1234 opened 2 hours ago

Shrad1234 commented 2 hours ago

Dear Authors,

I hope this message finds you well. I am currently working on research related to multimodal understanding of scenes and dialogue, and I require the dataset for this purpose. Upon reviewing the dataset, I noticed that the file named pro_seg_clip_train_ori.json appears to be missing. Additionally, if possible, could you please provide access to the features via Google Drive?

Thank you for your assistance. I look forward to your response.

Regards

patrick-tssn commented 2 hours ago

Thank you for your interest. Regarding the missing file, it was a temporary file with a typo that has now been fixed. You can run the preprocess.py smoothly. As for the video features, we cannot afford the cost of uploading such large data to Google Drive. However, we are exploring other platforms, such as AWS S3 or Aliyun. I will inform you as soon as we have a solution.