search

CurryYuan / X-Trans2Cap

[CVPR 2022] X-Trans2Cap: Cross-Modal Knowledge Transfer using Transformer for 3D Dense Captioning
Apache License 2.0
33 stars 3 forks source link

ScanNet dataset format for training the model #9

Closed EnnaSachdeva closed 1 year ago

EnnaSachdeva commented 2 years ago

Hi,

The training code seems to use these files for each of the ScanNet scenarios- "aligned_vert.npy, ins_label.npy, sem_label.npy, aligned_bbox.npy". But these files are missing from the repo. How is the original ScanNet data transformed into this format? Can you please share that codebase, or share the transformed data in the above format?

Thanks!!

CurryYuan commented 2 years ago

Hi, @EnnaSachdeva , please refer to Scan2Cap for data preparation.

1301358882 commented 7 months ago

Hi,

The training code seems to use these files for each of the ScanNet scenarios- "aligned_vert.npy, ins_label.npy, sem_label.npy, aligned_bbox.npy". But these files are missing from the repo. How is the original ScanNet data transformed into this format? Can you please share that codebase, or share the transformed data in the above format?

Thanks!!

Hi, have you solved it yet? How was it resolved?

CurryYuan commented 6 months ago

@1301358882 , I believe you can find what you need here.