AG-Net

Code for "Category-level 6D Pose Estimation Network Based on Attention Mechanism and Global Enhancement"

Requirements

The code has been tested with

• python 3.6.5
• pytorch 1.3.0
• CUDA 10.2

Other dependencies:

sh dependencies.sh

Data Processing

Please refer to work of Self-DPDN.

Evaluation

To test the model, please run:

python train.py --gpus 0 --dataset ${DATASET}

The string "DATASET" could be set as DATASET=REAL275 or DATASET=CAMERA25.

Results

Qualitative results on REAL275 test set:

Qualitative Comparison between FE-Net and VI-Net on the REAL275 Datasets.

Acknowledgements

Our implementation leverages the code from VI-Net,NOCS, DualPoseNet, and SPD.