Reflection-aware Generative Novel View Synthesis
Ref-GeNVS is a training-free method for reflection-consistent generative novel view synthesis that treats mirror images as two complementary views.
An arXiv paper proposes Ref-GeNVS, a training-free approach to generative novel view synthesis in scenes containing mirrors. It estimates the mirror plane, reflects camera poses to create virtual views, and applies mirror-gated attention plus reflection injection within a multi-view diffusion model. On synthetic and real mirror scenes, Ref-GeNVS outperforms recent generative NVS methods while requiring no fine-tuning.
- Treats each mirror image as two complementary views via estimated mirror plane
- Uses mirror-gated attention and reflection injection inside a multi-view diffusion backbone
- Training-free: inherits backbone generalizability without fine-tuning
- Outperforms recent generative NVS methods on synthetic and real mirror scenes
From the original article
1 min read at arxiv.orgWe propose Ref-GeNVS, a training-free, reflection-aware method for generative novel view synthesis (NVS) in mirror scenes. Existing multi-view diffusion models often fail to recognize the mirror in the scene and cannot exploit reflected content for scene generation. To fix this issue without additional training, our key idea is to treat a mirror image as two complementary views.
By GeonU Kim · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.