Novel view synthesis aims to generate high-quality unseen views from images at different viewpoints. However, ex isting methods often struggle to preserve fine details, lead ing to structural distortions in complex regions. In this pa per, we introduce a simple yet effective structure-aware ob jective function designed to enhance structural information in novel view synthesis. By leveraging the Structural Similarity Index (SSIM), our method attends to regions exhibiting significant structural distortions. We incorporate structural dissimilarity-based attention to highlight discrepancies in challenging regions between predicted and ground-truth images. It enables recent 3D scene representation models to achieve improved structural preservation, leading to more coherent representations. Experiments on synthetic and real world datasets demonstrate that our method enhances structural consistency, particularly in challenging regions.

Seungjae Lee is a master student at Graduate School of Advanced Imaging Science, Multimedia & Films (GSAIM), Chung-Ang University, where he is studying at Immersive Reality & Intelligent Systems Lab (IRIS LAB). Before joining Chung-Ang University, he received the B.S. degrees in School of Electrical Engineering at Myoungji University, Yongin, Korea, in 2024. His research interests include 3D Image Processing, Computer Vision, Artificial Intelligence.