PixGR is an efficient framework that integrates lightweight pixel-space generative modeling into feed-forward 3D Gaussian reconstruction to model scene uncertainty and maintains sub-second inference and remains competitive with substantially heavier pretrained generative models.
Abstract
We present PixGR, an efficient framework that integrates lightweight pixel-space generative modeling into feed-forward 3D Gaussian reconstruction to model scene uncertainty. Context observations are encoded once into a scene latent representation, which conditions generative decoding from noisy target views to produce a shared 3D Gaussian representation. We formulate this process using pixel-space flow matching, enabling end-to-end optimization of 3D reconstruction with direct pixel-space generative supervision through differentiable rendering. This avoids VAE-based latent compression and error propagation inherent in previous two-stage pipelines. To preserve feed-forward efficiency, PixGR caches the scene representation and performs iterative inference using a lightweight DiT decoder. Experiments demonstrate that PixGR outperforms regression-based approaches in challenging sparse-view settings, particularly with widely spaced observations and view extrapolation. Meanwhile, it maintains sub-second inference and remains competitive with substantially heavier pretrained generative models. Our project page is available at https://ranrhuang.github.io/pixgr/.
Sparse-view 4D reconstruction is highly under-constrained when only a few cameras are available. Generative priors can recover missing appearances, but video models are computationally impractical for iterative reconstruction, whereas single-view models produce temporally inconsistent pseudo-observations. Moreover, dir...
Bei-Zhen Zhao, Nan-Jie Yao, Bo-Yi Fu et al.· 0 citations
LucidLyra is a framework that makes generatively reconstructed 3D scenes directly queryable for semantics and instances in a feed-forward manner from video latents, and enables downstream applications including 3D scene editing, 3D asset mining, and robotic simulation.
Oliver Hahn, R. Marcuzzi, Sangeetha Grama Srinivasan et al.· 0 citations
Spatial intelligence requires scalable 3D environments that are both photorealistic and semantically grounded, yet existing indoor datasets are dominated by room-scale captures and costly LiDAR acquisition. Online real-estate platforms offer abundant panoramas of diverse multi-room homes, but sparse viewpoints and text...
Yuan-Yuan Gao, Zhi-Rui Liu, Yu-Ning Gong et al.· 0 citations
Zebra3D, a layerwise sparse-linear hybrid attention framework that improves the efficiency of pretrained 3D foundation models without training from scratch, achieves strong reconstruction and camera-pose accuracy at high attention sparsity, with training-free variants remaining close to their dense counterparts.
Utkarsh Saxena, Rui-Jing Yang, Mehdi Rezagholizadeh et al.· 0 citations
This work presents SplatFlux, a feed-forward latent-to-4DGS framework for both video-based 4D reconstruction and conditional 4D generation. Our key observation is that observed videos and conditionally generated content can be represented in the same pretrained video latent space, allowing both tasks to be formulated a...
Run-Min Zhang, Bin Ren, Zhu Yu et al.· 0 citations
OccSplat is introduced, a feed-forward framework whose core contribution is Occlusion-Aware Gaussian Densification, explicitly targeting missing scene geometry rather than missing pixels, to improve novel-view synthesis while retaining feed-forward inference without per-scene optimization.
Do-Tien Dung, K. D. Trung, K. Vu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.