Application of Neural Graphics Primitives Models for 3D Representation of Devastation Caused by Russian Aggression in Ukraine

Illia Oholtsov, Yuri G. Gordienko, Mariia Ladonia, Sergii Telenyk, Grzegorz Nowakowski, Sergii G. Stirenko

Abstract: This work investigates the feasibility of applying Neural Radiance Fields (NeRFs) for reconstructing 3D representations of damaged structures caused by the ongoing aggression of Russia against Ukraine. The drone footage depicting the devastation was utilized and three NeRF models, Instant-NGP, Nerfacto, and SplatFacto, were employed. The models were evaluated across various damage levels (0: no damage, 4: high damage) using visual quality metrics like Structural Similarity Index Measure (SSIM), Learned Perceptual Image Patch Similarity (LPIPS), Peak Signal-to-Noise Ratio (PSNR) and rendering speed metrics like frames per second (FPS) and the number of rays per second (NRS). All input data (videos frames) and evaluation results (rendered visualizations) are available as a Kaggle dataset (http://tiny.cc/srasxz). No clear correlation was observed between damage level and reconstruction quality metrics, suggesting these metrics might not be reliable indicators of damage severity. SplatFacto consistently achieved the highest rendering speed (FPS, NRS) and exhibited the best visual quality (SSIM, PSNR, LPIPS) across all damage levels. The findings suggest that NeRFs, particularly SplatFacto, hold promise for rapid reconstruction and visualization of damaged structures, potentially aiding in damage assessment, documentation, and cultural heritage preservation efforts. Moreover, the study sheds light on the potential applications of such advanced modeling techniques in archiving and documenting conflict zones, providing a valuable resource for future investigations, humanitarian efforts, and historical documentation. However, further research is needed to explore the generalizability and robustness of NeRFs in diverse real-world scenarios.