[question]test dataset groud truth: t-less/t-less_v2/test_primesense_gt_stats/

DLR-RM / AugmentedAutoencoder

Official Code: Implicit 3D Orientation Learning for 6D Object Detection from RGB Images

MIT License

341 stars 97 forks source link

[question]test dataset groud truth: t-less/t-less_v2/test_primesense_gt_stats/ #105

Closed qifengluo closed 2 years ago

qifengluo commented 2 years ago

In the dataset web, there is no the test_primesense_gt_stats (http://ptak.felk.cvut.cz/6DB/public/bop_datasets/) Every scense in test_primescense folder has gt.yaml, but it is not the same format with test code.

MartinSmeyer commented 2 years ago

The data format of the BOP challenge is different to what was used in the paper evaluation. The original data is unfortunately no longer available. If you want to compare, I would recommend using the BOP format as described here:

https://github.com/DLR-RM/AugmentedAutoencoder/tree/master#evaluate-a-model