Adversarial Camouflage in Autonomous Vehicles

Thesis notebooks for adversarial camouflage in autonomous vehicles (YOLOv8 + experiments).

Notebooks (open directly in Colab)

• final.ipynb – end-to-end workflow and results
  👉 https://colab.research.google.com/github/Sumanjali23/Adversarial-camouflage-in-Autonomous-Vehicles/blob/main/notebooks/final.ipynb

• YOLOv8_Stage3_Adversarial_Training.ipynb – training + adversarial experiments
  👉 https://colab.research.google.com/github/Sumanjali23/Adversarial-camouflage-in-Autonomous-Vehicles/blob/main/notebooks/YOLOv8_Stage3_Adversarial_Training.ipynb

How to run (Colab)

1. Open a notebook using one of the links above.
2. Run the setup cells to install dependencies.
3. Download the dataset (see below) and put it under data/.

Data (not stored in this repo)

Large datasets and weights are excluded from git (data/, runs/, weights/ are ignored).

Option A – Google Drive via gdown

!pip install -q gdown
DATA_ID = "1YzRfUeD6LOru8fdzQi4ElmkqpibxarEK"
!gdown {DATA_ID} -O data.zip
!unzip -q data.zip -d data/

Run locally (optional)

pip install -r requirements.txt