arXiv Computer Vision By Gengjia Chang, Ziteng Cui, Shuhong Liu

FujinSplat: Seeing Through Smoke with RAW-Domain Gaussian Splatting

Read the original on arXiv Computer Vision →

FujinSplat is a method for reconstructing clean 3D scenes from images captured through smoke by operating in the RAW image domain, where the camera’s ISP and the medium’s scattering effects remain separable. It first learns a per‑scene base ISP from hazy RAW captures, then uses a low‑dimensional correction space to fit per‑view corrections at training poses, and finally trains a scene‑agnostic controller to predict these corrections from RAW input. The corrected views supervise a single static 3D Gaussian representation, with a bounded per‑view residual to reconcile cross‑view inconsistencies, achieving superior performance on the RealX3D smoke benchmark compared to physics‑based and restoration‑then‑3DGS baselines.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Computer Vision
Sep 11

Tri-DehazeGS: Scene--Medium Decoupled Gaussian Splatting with Transmittance-Aware Optimization

Tri‑DehazeGS is a Gaussian Splatting framework that decouples clean scene reconstruction from atmospheric haze by representing the scene with Gaussian primitives and the haze medium with an independent view‑shared tri‑plane field. It uses a physical scattering model to compose hazy observations and introduces Medium‑Decoupled Transmittance Gradient Compensation (MD‑TGC) to re‑balance gradients in low‑transmittance regions without altering forward rendering. Experiments on real and synthetic haze benchmarks demonstrate that this approach improves clean novel‑view reconstruction.

By Kui Jiang, Yang Gu, Jiacheng Liu, Shiyu Liu, Youyu Chen, Hui Liu
arXiv Computer Vision
Aug 31

Manifold4D: Denoising on Point Cloud Rendered Manifolds for Video Re-shooting

Manifold4D introduces a new denoising strategy for video re‑shooting that injects a rendered point‑cloud directly into the initial noise manifold, eliminating the need for the render to be an explicit conditioning stream during denoising. This approach allows the network to rely solely on the source video as a visual condition, improving camera‑control accuracy on the DAVIS‑Traj benchmark and Vista4D set, with significant reductions in rotation and translation errors while maintaining video fidelity. User studies confirm enhanced trajectory following and dynamic consistency, especially for large yaw amplitudes and even when the render is corrupted.

By Yongqi Mao, Zijia Dai, Zhishuo Liu, Wei Xu, Kaiwei Wang, Guotao Meng
Hugging Face Trending Papers
Sep 10

Tri-DehazeGS: Scene--Medium Decoupled Gaussian Splatting with Transmittance-Aware Optimization

Tri-DehazeGS tackles the problem of reconstructing clean 3D scenes from hazy multi‑view images by decoupling the scene and the haze medium. It represents the scene with Gaussian primitives while modeling the haze as an independent view‑shared tri‑plane field, and composes hazy observations through a physical scattering model. The method introduces Medium‑Decoupled Transmittance Gradient Compensation (MD‑TGC) to balance gradients in low‑transmittance regions, leading to improved novel‑view reconstruction on real and synthetic haze benchmarks.