nihaomiao / CVPR23_LFDM

The pytorch implementation of our CVPR 2023 paper "Conditional Image-to-Video Generation with Latent Flow Diffusion Models"
BSD 2-Clause "Simplified" License
432 stars 43 forks source link

NATOPS Dataset preprocessing #22

Closed XiaoHaoPan closed 11 months ago

XiaoHaoPan commented 11 months ago

Hi, I want to train LFDM using NATOPS Dataset,but it(I downloaded it via your link) have a little different in preprocessing,it doesn't have json files.

it like this image

but this code image

nihaomiao commented 11 months ago

Hi, @XiaoHaoPan, thanks for your interest in our work! According to this line of code, it will create a JSON file and save it to the path json_path. So just change json_path to where you want to save the JSON file and set seg_txt_path to be the path of downloaded segmentation.txt. This JSON file will be used later.

XiaoHaoPan commented 11 months ago

Thank you for your help, the preprocessing process can seem time-consuming.