Abstract:Automatic generation of natural language from images has attracted extensive attention. In this paper, we take one step further to investigate generation of poetic language (with multiple lines) to an image for automatic poetry creation. This task involves multiple challenges, including discovering poetic clues from the image (e.g., hope from green), and generating poems to satisfy both relevance to the image and poeticness in language level. To solve the above challenges, we formulate the task of poem generation into two correlated sub-tasks by multi-adversarial training via policy gradient, through which the cross-modal relevance and poetic language style can be ensured. To extract poetic clues from images, we propose to learn a deep coupled visual-poetic embedding, in which the poetic representation from objects, sentiments and scenes in an image can be jointly learned. Two discriminative networks are further introduced to guide the poem generation, including a multi-modal discriminator and a poem-style discriminator. To facilitate the research, we have released two poem datasets by human annotators with two distinct properties: 1) the first human annotated image-to-poem pair dataset (with 8,292 pairs in total), and 2) to-date the largest public English poem corpus dataset (with 92,265 different poems in total). Extensive experiments are conducted with 8K images, among which 1.5K image are randomly picked for evaluation. Both objective and subjective evaluations show the superior performances against the state-of-the-art methods for poem generation from images. Turing test carried out with over 500 human subjects, among which 30 evaluators are poetry experts, demonstrates the effectiveness of our approach.

Image to Modern Chinese Poetry Creation via a Constrained Topic-aware Model

Poetry Generation Algorithm with Automatic Expansion of Keyword Semantic Information

Automated Thematic and Emotional Modern Chinese Poetry Composition

Generating Chinese Poetry from Images via Concrete and Abstract Information

How Images Inspire Poems: Generating Classical Chinese Poetry from Images with Memory Networks

Images2Poem: Generating Chinese Poetry from Image Streams.

Chinese Poetry Generation with a Working Memory Model

Poetic Expression Through Scenery: Sentimental Chinese Classical Poetry Generation from Images.

Chinese Poetry Generation with Metrical Constraints.

TPoet: Topic-Enhanced Chinese Poetry Generation.

Composing Like an Ancient Chinese Poet: Learn to Generate Rhythmic Chinese Poetry

Generating Thematic Chinese Poetry using Conditional Variational Autoencoders with Hybrid Decoders

A new context-aware approach for automatic Chinese poetry generation

Poetry2Image: An Iterative Correction Framework for Images Generated from Chinese Classical Poetry

Beyond Narrative Description: Generating Poetry from Images by Multi-Adversarial Training

Generating Chinese Classical Poems Based on Images

A New Automatic Chinese Poetry Generation Model Based on Neural Network

A Multi-Modal Chinese Poetry Generation Model

Chinese Traditional Poetry Generating System Based on Deep Learning

Can Machine Generate Traditional Chinese Poetry? A Feigenbaum Test

Semi-supervised Chinese Poem-to-Painting Generation via Cycle-consistent Adversarial Networks