This repository contains implementation/solution to the following Computer Vision assignment:

Problem: The goal is to detect spaceships which have been fitted with a cloaking device that makes them less visible. You are expected to use a deep learning model to complete this task. The model will take a single channel image as input and detects the spaceship (if it exists). Not all image will contain a spaceship, but they will contain no more than 1. For any spaceship, the model should predict their bounding box and heading. This can be described using five parameters:

• X and Y position (centre of the bounding box)
• Yaw (direction of heading)
• Width (size tangential to the direction of yaw)
• Height (size along the direct of yaw)

A sample spaceship image is given in example.png.

The metric for the model is AP at an IOU threshold of 0.7, for at least 1000 random samples, with the default generation parameters (see main.py).

Evaluation Criteria:

• Model metric, score as high as you can while being under 2 million trainable parameters. The average submitted AP for this assignment is 0.7.
• Model architecture
• Loss function
• Code readability and maintainability, please follow general python conventions

Deliverables

1. Final score reported in final_score.txt
2. A summary of the model architecture in architecture_summary.txt
3. A train.py script that allows the same model to be reproduced
4. The final model weights in model.pth.tar
5. A requirements.txt file that includes all python dependencies and their versions
6. A main.py file that reproduces the reported score

In addition to those in requirements.txt, following dependencies are needed

cudatoolkit 10.2
cuda 10.1
gcc 6.3.0

First, compile the CUDA extension. (not needed for evaluation)

cd cuda_op
python setup.py install

Then, run train.py

cd ..
python train.py

plot_examples and parts of helpers.py (code for generating spaceship images) were given by the assignment.

The code for GIoU loss is copied from https://github.com/lilanxiao/Rotated_IoU, and is contained in directory cuda_op and files box_intersection_2d.py, min_enclosing_box.py, oriented_iou_loss.py, utiles.py.

main.py, network.py, train.py, and rest of helpers.py are implemented from scratch.