![](/project/quobilenet/featured_hucfdae3a040d4db0c83ba4bffc2937ac5_438485_c340649787cfda61cd2df6744a48366e.webp)
Our goal was to design a MobileNetV2-based Hybrid Quantum-Classical Object Detector. To start off, we built a CNN model (SimpleNet) whose classical version achieves an average accuracy of 99.60% on a 3-class classification problem using numbers 0, 6, and 9 from the MNIST dataset. We replaced one of the 4 convolutional layers with a quantum equivalent: a quanvolutional layer, to create QuanvNet and compared its performance to SimpleNet. To learn more, please visit our GitHub page.