First of all, thank you very much for your valuable research, and I have some questions about Ted dataset.
In the paper, I found that Ted is preprocessed by cropping the upper part of the human body from the videos. I will be appreciate to know the detail way to crop Ted.
When I run the commend of "python load_videos.py --metadata ../data/ted384-metadata.csv --format .mp4 --out_folder ../data/TED384-v2 --workers 8 --image_shape 384,384", I only find the ted-metadata.csv instead of ted384-metadata.csv in data folder. And I have some qusetions about the ted-metadata.csv. First, the classes of start and end has all zero number. Second, the class of partition gets all "train". Is the ted-metadata.csv in data folder right? if not, can you provide the right ted384-metadata.csv?
Again, Thank you for your remarkble work!
First of all, thank you very much for your valuable research, and I have some questions about Ted dataset.