Feature Field Fusion for few-shot novel view synthesis

Junting Li, Yanghong Zhou, Jintu Fan, Dahua Shou, Sa Xu, P. Y. Mok

Abstract: Highlights•A novel NeRF method using spatial feature maps to improve sparse view synthesis.•Concurrently learning both feature & color fields with pixel-level loss supervisions.•A fusion scheme regularizing color prediction with learned features as condition guide.•Extensive tests validate the performance and model-agnostic advantages of the method.