7 papers
A Practical Investigation of Spatially-Controlled Image Generation with Transformers
Guoxuan Xia, Harleen Hanspal, Petru-Daniel Tudosiu +2
Enabling image generation models to be spatially controlled is an important area of research, empowering users to better generate images according to their own fine-grained specifi…
Adapting Vision-Language Models for Evaluating World Models
Mariya Hendriksen, Tabish Rashid, David Bignell +5
World models - generative models that simulate environment dynamics conditioned on past observations and actions - are gaining prominence in planning, simulation, and embodied AI.…
Exploiting Mixture-of-Experts Redundancy Unlocks Multimodal Generative Abilities
Raman Dutt, Harleen Hanspal, Guoxuan Xia +5
In this work, we undertake the challenge of augmenting the existing generative capabilities of pre-trained text-only large language models (LLMs) with multi-modal generation capabi…
Generating Compositional Scenes via Text-to-image RGBA Instance Generation
Alessandro Fontanella, Petru-Daniel Tudosiu, Yongxin Yang +2
Text-to-image diffusion generative models can generate high quality images at the cost of tedious prompt engineering. Controllability can be improved by introducing layout conditio…
Improving Object Detection via Local-global Contrastive Learning
Danai Triantafyllidou, Sarah Parisot, Ales Leonardis +1
Visual domain gaps often impact object detection performance. Image-to-image translation can mitigate this effect, where contrastive approaches enable learning of the image-to-imag…
MULAN: A Multi Layer Annotated Dataset for Controllable Text-to-Image Generation
Petru-Daniel Tudosiu, Yongxin Yang, Shifeng Zhang +5
Text-to-image generation has achieved astonishing results, yet precise spatial controllability and prompt fidelity remain highly challenging. This limitation is typically addressed…
Optimisation-Based Multi-Modal Semantic Image Editing
Bowen Li, Yongxin Yang, Steven McDonagh +3
Image editing affords increased control over the aesthetics and content of generated images. Pre-existing works focus predominantly on text-based instructions to achieve desired im…