Log In Sign Up

Shared Latent Space of Font Shapes and Impressions

by   Jihun Kang, et al.

We have specific impressions from the style of a typeface (font), suggesting that there are correlations between font shape and its impressions. Based on this hypothesis, we realize a shared latent space where a font shape image and its impression words are embedded in a cross-modal manner. This latent space is useful to understand the style-impression correlation and generate font images by specifying several impression words. Experimental results with a large style-impression dataset prove that it is possible to accurately realize the shared latent space, especially for shape-relevant impression words, and then use the space to generate font images with various impressions.


page 1

page 2

page 3

page 4


Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?

We propose an efficient algorithm to embed a given image into the latent...

Deep Latent Space Learning for Cross-modal Mapping of Audio and Visual Signals

We propose a novel deep training algorithm for joint representation of a...

Visualizing Missing Surfaces In Colonoscopy Videos using Shared Latent Space Representations

Optical colonoscopy (OC), the most prevalent colon cancer screening tool...

Decoding a Neural Retriever's Latent Space for Query Suggestion

Neural retrieval models have superseded classic bag-of-words methods suc...

Manifold Relevance Determination

In this paper we present a fully Bayesian latent variable model which ex...

Unsupervised physics-informed disentanglement of multimodal data for high-throughput scientific discovery

We introduce physics-informed multimodal autoencoders (PIMA) - a variati...

LaLaLoc: Latent Layout Localisation in Dynamic, Unvisited Environments

We present LaLaLoc to localise in environments without the need for prio...