News
Improving Image Embeddings with Colour Features in Indoor Scene Geolocation
Congratulations to Opeyemi Bamigbade and co-authors Mark Scanlon, and John Sheppard on the publication of Improving Image Embeddings with Colour Features in Indoor Scene Geolocation in IEEE Access.
Co-authors: Mark Scanlon, and John Sheppard.
AI-generated summary of the contribution: The authors propose a novel model architecture that enhances image embeddings with colour features to improve image geolocation in indoor scenes. By integrating N-dominant colours and colour histogram vectors with image embedding from deep metric learning and classification perspectives, the proposed model achieves competitive performance on the Hotels-50K dataset. The results indicate that the integration of colour features improves image embedding, surpassing the performance of using embedding alone. The classification approach yields higher accuracy compared to deep metric learning methods. The findings have implications for the design of more robust image geolocation systems, particularly in indoor environments.