3 papers
cs.CV2026
Knowledge Visualization: A Benchmark and Method for Knowledge-Intensive Text-to-Image Generation
Ran Zhao, Sheng Jin, Size Wu +5
Recent text-to-image (T2I) models have demonstrated impressive capabilities in photorealistic synthesis and instruction following. However, their reliability in knowledge-intensive…
cs.CV2025
SA-LUT: Spatial Adaptive 4D Look-Up Table for Photorealistic Style Transfer
Zerui Gong, Zhonghua Wu, Qingyi Tao +2
Photorealistic style transfer (PST) enables real-world color grading by adapting reference image colors while preserving content structure. Existing methods mainly follow either ap…
cs.CV2025
OpenUni: A Simple Baseline for Unified Multimodal Understanding and Generation
Size Wu, Zhonghua Wu, Zerui Gong +5
In this report, we present OpenUni, a simple, lightweight, and fully open-source baseline for unifying multimodal understanding and generation. Inspired by prevailing practices in…