Pith. sign in

REVIEW 11 cited by

pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2312.12337 v4 pith:KZ2ZKILR submitted 2023-12-19 cs.CV cs.LG

pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction

classification cs.CV cs.LG
keywords gaussiandistributionfieldmodelpairspixelsplatprobabilityradiance
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original 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 editable 3D radiance field.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction

    cs.CV 2026-05 unverdicted novelty 7.0

    GenRecon lifts object-level generative priors to scene-scale reconstruction by chunking scenes and using projection-based conditioning on multi-view features, claiming 16% better results than prior methods.

  2. No Pose, No Problem in 4D: Feed-Forward Dynamic Gaussians from Unposed Multi-View Videos

    cs.CV 2026-05 unverdicted novelty 7.0

    NoPo4D is the first feed-forward system for dynamic 4D Gaussian splatting from unposed multi-view videos, using velocity decomposition supervised by optical flow and a bidirectional motion encoder.

  3. PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation

    cs.CV 2026-07 conditional novelty 6.0

    PixGS is a single-stage pixel-space diffusion model that directly produces high-quality 3D Gaussian Splats from text or images in ~1s, outperforming multi-stage latent methods on standard benchmarks.

  4. PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation

    cs.CV 2026-07 unverdicted novelty 6.0

    A single-stage pixel-space diffusion model for direct 3D Gaussian Splat generation that bypasses latent compression and adds geometric supervisions to outperform prior multi-stage methods.

  5. Learning Efficient 4D Gaussian Representations from Monocular Videos with Flow Splatting

    cs.CV 2026-06 unverdicted novelty 6.0

    Flow Splatting extends 4D Gaussian volumes with time-varying means and covariances, approximates a velocity field, and splats it to render optical flow for supervising dynamic reconstruction from monocular video.

  6. Life After Benchmark Saturation: A Case Study of CORE-Bench

    cs.AI 2026-06 unverdicted novelty 6.0

    Using CORE-Bench as a case study, the paper shows that saturated benchmarks can still deliver insights on efficiency, reliability, model-scaffold differences, and human collaboration even after accuracy plateaus, and ...

  7. Self-Learning Expression Deformations for Data-Efficient Gaussian Avatars

    cs.CV 2026-06 unverdicted novelty 6.0

    SAGE self-learns Gaussian expression deformations via joint surfel-SDF optimization and self-supervised consistency, enabling comparable avatar quality from single frames, monocular rotations, or one-shot inputs.

  8. GaussianGPT: Towards Autoregressive 3D Gaussian Scene Generation

    cs.CV 2026-03 conditional novelty 6.0

    A causal transformer with 3D RoPE generates vector-quantized 3D Gaussian latent grids autoregressively, enabling unconditional synthesis, completion, and open-ended outpainting of indoor scenes.

  9. CamCo: Camera-Controllable 3D-Consistent Image-to-Video Generation

    cs.CV 2024-06 unverdicted novelty 6.0

    CamCo equips image-to-video generators with Plücker-coordinate camera inputs and epipolar attention to improve 3D consistency and camera controllability.

  10. VistaVLA: Geometry- and Semantic-Aware 3D Gaussian-Grounded VLA for Robotic Manipulation

    cs.RO 2026-07 conditional novelty 5.5

    VistaVLA lifts multi-view vision-language features into 3D Gaussians, compresses them 99% via Merge-then-Query, and improves real-robot manipulation success by ~23% over baselines.

  11. VistaVLA: Geometry- and Semantic-Aware 3D Gaussian-Grounded VLA for Robotic Manipulation

    cs.RO 2026-07 conditional novelty 5.0

    VistaVLA lifts 2D semantic features into 3D Gaussians and compresses them to 64 tokens, improving robot manipulation success by 22.8% over a 2D VLA baseline in real-world tasks.