Bimodal Camera Pose Prediction for Endoscopy

The data from our paper and sequences from additional anatomies can be accessed for the SimCol-to-3D MICCAI 2022 challenge.

⚙ Getting started

The full dataset is now publicly available here. Access through Synapse is no longer required. To visualize the data like above run visualize_3D_data.py with the updated path to the data.

🌍 Submitting to the SimCol-to-3D MICCAI challenge

The SimCol Challenge is now closed. The results can be found in the challenge paper here.

📊 Evaluation and challenge leader board

Tasks 1, 2, and 3 will be evaluated according to the scripts in evaluation. To evaluate your method on validation data, generate predictions according to the docker_templates and run

python evaluation/eval_synthetic_depth.py /path/to/predictions /path/to/groundtruth