DeepFont is a new CUDA-based system by researchers from Adobe, Google, and Snapchat that achieves higher than 80% top-5 accuracy. Graphic designers have the desire to identify the fonts they encounter in daily life for later use. While they might take a photo of the text of a particularly interesting font and seek out an expert to identify the font, the manual identification process is extremely tedious and error-prone.
Several websites allow users to search and recognize fonts by font similarity, including Identifont, MyFonts, WhatTheFont, and Fontspring. All of them rely on tedious humans interactions and high-quality manual pre-processing of images, and the accuracies are still unsatisfactory.
In comes the Deep Font system. Using CUDA, GeForce and a Convolutional Neural Network (CNN), the system achieves an accuracy of higher than 80% (top-5) on their collected dataset, and also produces a good font similarity measure for font selection and suggestion.
Read the entire research paper >>
Font Identification From Real-World Images
Sep 18, 2015
Discuss (0)
Related resources
- DLI course: Deep Learning for Industrial Inspection
- GTC session: Simplifying OCR Serving with Triton Inference Server
- GTC session: Tencent Text-to-Image Generative Model
- GTC session: Flexible Text-to-Parameter Translation for 3D Game Character Auto-Creation
- NGC Containers: quickstart-rapidsai
- NGC Containers: MATLAB