CVJun 1, 2021

Language-Driven Image Style Transfer

arXiv:2106.00178v366 citations
Originality Incremental advance
AI Analysis

This addresses a creativity and accessibility issue in visual effect applications by enabling more natural, controllable style transfer through language instructions.

The paper tackles the problem of style transfer requiring pre-prepared style images by introducing language-driven artistic style transfer (LDAST), where text guides style manipulation, and proposes CLVA to achieve this, showing effective and superb results in experiments.

Despite having promising results, style transfer, which requires preparing style images in advance, may result in lack of creativity and accessibility. Following human instruction, on the other hand, is the most natural way to perform artistic style transfer that can significantly improve controllability for visual effect applications. We introduce a new task, language-driven artistic style transfer (LDAST), to manipulate the style of a content image, guided by a text. We propose contrastive language visual artist (CLVA) that learns to extract visual semantics from style instructions and accomplish LDAST by the patch-wise style discriminator. The discriminator considers the correlation between language and patches of style images or transferred results to jointly embed style instructions. CLVA further compares contrastive pairs of content images and style instructions to improve the mutual relativeness. The results from the same content image can preserve consistent content structures. Besides, they should present analogous style patterns from style instructions that contain similar visual semantics. The experiments show that our CLVA is effective and achieves superb transferred results on LDAST.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes