← Back to projects

HEDWIG: Learning Geospatial Embeddings for Large-Scale Retrieval

One-line summary: Built a ViCLIP-based geolocation system that learns richer geospatial embeddings from multi-frame panoramic imagery and captions.

Key Results

What I Built

Technical Approach

Key Insight

Averaging embeddings across viewpoints weakens location signal; weighted multi-frame representations improve geospatial retrieval and clustering.

Tools / Models Used

Python, PyTorch, ViCLIP, CLIP, geospatial clustering, retrieval, similarity search.

Optional links to paper / GitHub / PDF