search

henghuiding / MeViS

[ICCV 2023] MeViS: A Large-scale Benchmark for Video Segmentation with Motion Expressions
https://henghuiding.github.io/MeViS/
MIT License
472 stars 19 forks source link

Mask GroundTruth #4

Closed ChenHsing closed 9 months ago

ChenHsing commented 9 months ago

I download the mask_dict.json from the google cloud, however, there seems to be some Encoding error， could you please provide a Annotation folder such as Youtube-vos and Davis?

henghuiding commented 9 months ago

Hi,

Thank you for your attention.

Please kindly verify the SHA256 sum of the mask_dict.json files to ensure the data intergrity:

f9d9b01a2a6d7a004e2e7a664ce2b9b1576218e975030f8e4df4f66111fb253f train/mask_dict.json
6e711a3f713cd95a825f536323e434b4d80f85a8e4e66d0cd662408e8779db50 valid_u/mask_dict.json

There should be no encoding error when using our code.

Additionally, we also provide ReferFormer_dataloader.py for reading .json, with which you can train ReferFormer on MeViS dataset.

ChenHsing commented 9 months ago

But could you please provide a script to convert the mask_dict.json to mask annotation with jpg？ I have some trouble in this step.