Illustration2vec
Visit Toolillustration2vec is a deep learning library for estimating tags and extracting semantic feature vectors from illustrations. It provides pre-trained models for tag prediction and feature extraction.
At a glance
Trending
illustration2vec is a deep learning library for estimating tags and extracting semantic feature vectors from illustrations. It provides pre-trained models for tag prediction and feature extraction.
Trending
About
illustration2vec is a simple deep learning library designed for estimating a set of tags and extracting semantic feature vectors from given illustrations. It leverages Convolutional Neural Networks and offers pre-trained models for immediate use. The library supports both tag prediction, classifying tags into general, copyright, character, and rating categories, and the extraction of 4,096-dimensional real or binary feature vectors. It requires Python libraries like numpy, scipy, PIL/Pillow, skimage, and either Caffe or Chainer. The tool is open-source, with models provided under the MIT License, making it suitable for researchers and developers in image analysis and machine learning.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending