Object detection using Res-YoLo
Applied the model architecture taken from the paper, onto an open source car detection Dataset.
- Uses a ResNet backbone as a feature Extracter, and uses a YoLo like architecture to detect cars.
- Fine-Tuned the architecture on a car detection dataset, by freezing the intial few ResNet layers and training the custom detector in a regression like fashion.
- Tech stack: Pytorch, numpy, matplotlib