arXiv:2509.17329v2 Announce Type: replace 
Abstract: Smoke in real-world scenes can severely degrade image quality and hamper visibility. Recent image restoration methods either rely on data-driven priors that are susceptible to hallucinations, or are limited to static low-density smoke. We introduce SmokeSeer, a method for simultaneous 3D scene reconstruction and smoke removal from multi-view video sequences. Our method uses thermal and RGB images, leveraging the reduced scattering in thermal images to see through smoke. We build upon 3D Gaussian splatting to fuse information from the two image modalities, and decompose the scene into smoke and non-smoke components. Unlike prior work, SmokeSeer handles a broad range of smoke densities and adapts to temporally varying smoke. We validate our method on synthetic data and a new real-world smoke dataset with RGB and thermal images. We provide an open-source implementation and data on the project website.

تم تقديم SmokeSeer كطريقة جديدة لإزالة الدخان وإعادة بناء المشاهد ثلاثية الأبعاد من تسلسلات الفيديو متعددة الزوايا، باستخدام صور حرارية وRGB لإدارة كثافات الدخان المتغيرة بشكل فعال. تستفيد هذه الطريقة من التشتت المنخفض في الصور الحرارية، مما يسمح برؤية أوضح من خلال الدخان، وهو تقدم كبير في تكنولوجيا استعادة الصور.

SmokeSeer se ha presentado como un método novedoso para la eliminación de humo y la reconstrucción 3D de escenas a partir de secuencias de video multivista, utilizando imágenes térmicas y RGB para gestionar eficazmente las densidades de humo variables. Este enfoque aprovecha la dispersión reducida en las imágenes térmicas, permitiendo una mejor visibilidad a través del humo, lo que representa un avance significativo en la tecnología de restauración de imágenes.

SmokeSeer a été introduit comme une méthode novatrice pour l'élimination de la fumée et la reconstruction 3D de scènes à partir de séquences vidéo multi-vues, utilisant à la fois des images thermiques et RGB pour gérer efficacement les densités de fumée variables. Cette approche tire parti de la diffusion réduite dans les images thermiques, permettant une meilleure visibilité à travers la fumée, ce qui constitue une avancée significative dans la technologie de restauration d'images.

SmokeSeer has been introduced as a novel method for smoke removal and 3D scene reconstruction from multi-view video sequences, utilizing both thermal and RGB images to effectively manage varying smoke densities. This approach leverages reduced scattering in thermal images, allowing for clearer visibility through smoke, which is a significant advancement in image restoration technology.

SmokeSeer: 3D Gaussian Splatting for Smoke Removal and Scene Reconstruction

Was this article worth reading? Share it

SwapAnything.io

C-THERMAL

Open Source Surveillance

VSDECO

Chroma

Fakeface

Ready to build your own newsroom?