Accurate Prediction of Bounding Boxes in Object Detection

Introduction

This repository contains experiments to reproduce the results of the paper Bounding Box Regression with Uncertainty for Accurate Object Detection (He et al.). For more details, check out this report / slides / presentation.

Running Experiments

• Run the notebook file kl_loss_MMDet.ipynb with CPU configuration and follow the instructions in the file.

• Run the notebook file kl_loss_pytorch.ipynb and follow the instructions in the file. Note that Google Drive has been linked to the notebook for data and output storage.

Visualization

See a visualization of the objects getting detected using this app.

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

MIT