FULL PORTFOLIO
January - February 2019

Image Categorisation

As part of a university module, a peer and I pursued a bag-of-visual-words approach to the multi-class image categorisation problem, using a subset of the Caltech 101 dataset.

We used dense SIFT to obtain descriptors. We experimented with K-means clustering and Random Forests for codebook creation. We again used Random Forests for classification.

We discussed our findings in a report. All relevant code can be found in the associated GitHub repository.