search

ufal / MLASK

EACL 2023 paper "MLASK: Multimodal Summarization of Video-based News Articles"
https://aclanthology.org/2023.findings-eacl.67
Apache License 2.0
10 stars 0 forks source link

Video and Image feature extraction dimensions #4

Closed willxxy closed 1 year ago

willxxy commented 1 year ago

Hello,

For the video and image features (i am using ig65m and vit, respectively), what is the format/dimensionality of the .npy files?

mateuk commented 1 year ago

Hi! Please take a look at the notebooks in MLASK/feature_extraction and the examples of using np.save() - they correspond to the format expected by dataloader in MLASK/src/data/data_loader.py.