Hyperbolic Image Embeddings

04/03/2019
by   Valentin Khrulkov, et al.
26

Computer vision tasks such as image classification, image retrieval and few-shot learning are currently dominated by Euclidean and spherical embeddings, so that the final decisions about class belongings or the degree of similarity are made using linear hyperplanes, Euclidean distances, or spherical geodesic distances (cosine similarity). In this work, we demonstrate that in many practical scenarios hyperbolic embeddings provide a better alternative.

READ FULL TEXT
research
03/15/2022

On Hyperbolic Embeddings in 2D Object Detection

Object detection, for the most part, has been formulated in the euclidea...
research
03/29/2021

von Mises-Fisher Loss: An Exploration of Embedding Geometries for Supervised Learning

Recent work has argued that classification losses utilizing softmax cros...
research
09/18/2023

Hyperbolic vs Euclidean Embeddings in Few-Shot Learning: Two Sides of the Same Coin

Recent research in representation learning has shown that hierarchical d...
research
09/01/2022

Spherical Graph Drawing by Multi-dimensional Scaling

We describe an efficient and scalable spherical graph embedding method. ...
research
03/11/2022

Hyperbolic Image Segmentation

For image segmentation, the current standard is to perform pixel-level o...
research
09/04/2023

Interactive Design and Optics-Based Visualization of Arbitrary Non-Euclidean Kaleidoscopic Orbifolds

Orbifolds are a modern mathematical concept that arises in the research ...
research
06/30/2023

Sphere2Vec: A General-Purpose Location Representation Learning over a Spherical Surface for Large-Scale Geospatial Predictions

Generating learning-friendly representations for points in space is a fu...

Please sign up or login with your details

Forgot password? Click here to reset