diff --git a/README.md b/README.md index 1f21e4c15b2a95a66397fcf97e79b0f839aaac60..9597f187726c4d00f8aa8af72b68bdc248a11e64 100644 --- a/README.md +++ b/README.md @@ -25,6 +25,8 @@ Link the datasets in the following folders with symlink. ├── ycbv ``` +The pretrained models are available on [onedrive](https://mailstsinghuaeducn-my.sharepoint.com/:f:/g/personal/liuxy21_mails_tsinghua_edu_cn/EgT-saAOPtROonp9o_nmPDQBeIO2x7RjvhMSXhMMTMgrlg?e=L2bX2f) (password: groupji). + ## Training We train a model separately for each of the BOP datasets. To train the TUDL refinement model, run ``` @@ -35,11 +37,13 @@ To train a refinement model on other BOP datasets, replace `ycbv` with one of th ## Testing 1. Generate GDRN results with save_gdrn.sh and put the trained RAFT models in folder "model_weights". + 2. Run ``` ./test_fast.sh <DATASET> <RESULT_PATH> <GPU_ID> ``` Use "test_fast_pbr.sh" for PBR results of YCB-V, T-LESS and TUD-L. + 3. Run ``` python -m additional_scripts.convert_result --tar_dir <TAR_DIR> --dataset <DATASET>