Lee, Seokyeong Choi, Junyong Kim, Seungryong Kim, Ig-Jae Cho, Junghyun 2025-10-30T07:00:43Z 2025-10-30T07:00:43Z 2025-10-30 2025-09 https://pubs.kist.re.kr/handle/201004/153372 Neural Radiance Fields (NeRF) have demonstrated strong performance in novel view synthesis under idealized conditions such as dense multi-view observations, consistent illumination, and known camera poses. However, these assumptions often do not hold in real-world scenarios, where inputs are sparse and lighting varies significantly. This paper presents a novel regularization framework for few-shot NeRF reconstruction under unconstrained illumination. By leveraging intrinsic, illumination-invariant representations (e.g., albedo), our method enforces cross-view appearance consistency, leading to more stable synthesis. To further improve applicability, we propose a lightweight variant that achieves comparable improvements with significantly reduced computational cost. We also establish new benchmarks that reflect diverse illumination and viewpoint conditions. Extensive experiments show that our method improves robustness and rendering quality across challenging real-world scenes, without relying on dense inputs or manual supervision. English Institute of Electrical and Electronics Engineers Inc. Leveraging Intrinsic Components for Few-Shot Neural Radiance Fields in Unconstrained Illumination Article 10.1109/ACCESS.2025.3610908 1 IEEE Access, v.13, pp.169150 - 169165 IEEE Access 13 169150 169165 Y scie scopus 001587270100020 2-s2.0-105016521227 Computer Science, Information Systems Engineering, Electrical & Electronic Telecommunications Computer Science Engineering Telecommunications Article VIEW SYNTHESIS RECONSTRUCTION Color Rendering (computer graphics) Image color analysis Geometry Benchmark testing Robustness Optimization Cameras Illumination decomposition neural radiance fields view synthesis Neural radiance field Lighting