Landmark Recognition Using VGG16 Training
Ruchi Jha, Prerna Jain, Sandeep Tayal, Ashish Arunkumar Sharma · 2021
The paper proposes a solution to Google landmark recognition challenge 2019 hosted on Kaggle and an intuition for facilitating a tour guide recommender engine and visualization. Too often while surfing our photos from last trip we stumble upon a few which leave us clueless about the castle or the river behind us. A landmark recognition engine can be the best anodyne. The challenge comprised a huge and vast dataset classified into sheer amount of predictions. With the presence of ample amount of “junk” images, broken links and images with multi-class predictions, our method attempts to predict landmarks using CNN, pre-trained with VGG16 neural network used with transfer learning from ImageNet.