PixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction
Citations Over TimeTop 1% of 2024 papers
Abstract
We introduce pixelSplat, a feed-forward model that learns to reconstruct 3D radiance fields parameterized by 3D Gaussian primitives from pairs of images. Our model features real-time and memory-efficient rendering for scalable training as well as fast 3D reconstruction at inference time. To overcome local minima inherent to sparse and locally supported representations, we predict a dense probability distribution over 3D and sample Gaussian means from that probability distribution. We make this sampling operation differentiable via a reparameterization trick, allowing us to back-propagate gradients through the Gaussian splatting representation. We benchmark our method on wide-baseline novel view synthesis on the real-world RealEstate10k and ACID datasets, where we outperform state-of-the-art light field transformers and accelerate rendering by 2.5 orders of magnitude while reconstructing an interpretable and ed-itable 3D radiance field. Additional materials can be found on the project website.11dcharatan.github.io/pixelsplat
Related Papers
- → Scalability Issues of Blockchain Technology(2020)30 cited
- → INVITED PAPER: Scalability and Performance Issues in Deeply Embedded Sensor Systems(2009)7 cited
- → On the scalability of multistage interconnection networks(2004)3 cited
- → Using Empirical Data for Scalability Analysis of Parallel Applications(2019)1 cited
- RESEARCH ON THE SCALABILITY OF THE LARGE SCALE PARALLEL APPLICATION PROGRAMS(2000)