Timezone: »

Lightweight Generative Adversarial Networks for Text-Guided Image Manipulation
Bowen Li · Xiaojuan Qi · Philip Torr · Thomas Lukasiewicz

Wed Dec 09 09:00 AM -- 11:00 AM (PST) @ Poster Session 3 #932

We propose a novel lightweight generative adversarial network for efficient image manipulation using natural language descriptions. To achieve this, a new word-level discriminator is proposed, which provides the generator with fine-grained training feedback at word-level, to facilitate training a lightweight generator that has a small number of parameters, but can still correctly focus on specific visual attributes of an image, and then edit them without affecting other contents that are not described in the text. Furthermore, thanks to the explicit training signal related to each word, the discriminator can also be simplified to have a lightweight structure. Compared with the state of the art, our method has a much smaller number of parameters, but still achieves a competitive manipulation performance. Extensive experimental results demonstrate that our method can better disentangle different visual attributes, then correctly map them to corresponding semantic words, and thus achieve a more accurate image modification using natural language descriptions.

Author Information

Bowen Li (University of Oxford)
Xiaojuan Qi (The University of Hong Kong)
Philip Torr (University of Oxford)
Thomas Lukasiewicz (University of Oxford)

More from the Same Authors