Conditional plane-based multi-scene representation for novel view synthesis

Uchitha Rajapaksha, Hamid Laga, Dean Diepeveen, Mohammed Bennamoun, Ferdous Sohel

Abstract: Highlights•We present a novel multi-scene representation that uses twelve explicit feature planes.•The method uses encoder-less generalization to learn discriminative features for each scene.•We represent multiple dynamic scenes without relying on optical flow estimation.•The auto-decoded latent (the scene dimension) can interpolate between scenes.•It achieves state-of-the-art rendering results for both static and dynamic scenes.

External IDs:doi:10.1016/j.neucom.2025.131657