Search:
Match:
2 results

Analysis

This paper introduces a novel Driving World Model (DWM) that leverages 3D Gaussian scene representation to improve scene understanding and multi-modal generation in driving environments. The key innovation lies in aligning textual information directly with the 3D scene by embedding linguistic features into Gaussian primitives, enabling better context and reasoning. The paper addresses limitations of existing DWMs by incorporating 3D scene understanding, multi-modal generation, and contextual enrichment. The use of a task-aware language-guided sampling strategy and a dual-condition multi-modal generation model further enhances the framework's capabilities. The authors validate their approach with state-of-the-art results on nuScenes and NuInteract datasets, and plan to release their code, making it a valuable contribution to the field.
Reference

Our approach directly aligns textual information with the 3D scene by embedding rich linguistic features into each Gaussian primitive, thereby achieving early modality alignment.

DreamOmni3: Scribble-based Editing and Generation

Published:Dec 27, 2025 09:07
1 min read
ArXiv

Analysis

This paper introduces DreamOmni3, a model for image editing and generation that leverages scribbles, text prompts, and images. It addresses the limitations of text-only prompts by incorporating user-drawn sketches for more precise control over edits. The paper's significance lies in its novel approach to data creation and framework design, particularly the joint input scheme that handles complex edits involving multiple inputs. The proposed benchmarks and public release of models and code are also important for advancing research in this area.
Reference

DreamOmni3 proposes a joint input scheme that feeds both the original and scribbled source images into the model, using different colors to distinguish regions and simplify processing.