Note - there is no dataset included due to space constraints

1. Clone repo

git clone [email protected]:baj31415/visual-place-recognition.git \
&& cd visual-place-recognition

2. Build

mkdir build && cd build
cmake ..
make 

3. Run executable, which should generate a html file for the the most similar images in the dataset to the query image

./main

4. Open the webpage in a web browser

• Sift Descriptors for each image in the dataset
• Kmeans clustering of the Sift descriptors
• Bag of Visual Words Dictionary
• Histogram Computation
• Concatenation of Histograms from the Dataset
• TF-IDF on the Histograms
• Histogram comparison with cosine similarity