MoMA
Multi-modal LLM for image personalization
Generate large images on pre-trained Stable Diffusion models
Rerun viewer with Gradio
Zero-Shot Material Transfer from a Single Image
Meta Llama3 8b with Llava Multimodal capabilities
Unbounded Sparse-view Pose-free Gaussian Splatting in 40s
Bring song ideas to life
High-fidelity Virtual Try-on