Pith. sign in

REVIEW 5 cited by

MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2411.04924 v1 pith:T7EZW5NM submitted 2024-11-07 cs.CV

Yuedong Chen , Chuanxia Zheng , Haofei Xu , Bohan Zhuang , Andrea Vedaldi , Tat-Jen Cham , Jianfei Cai This is my paper
classification cs.CV
keywords mvsplat360viewsmodelfeed-forwardsparsevideovisualbenchmark
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We introduce MVSplat360, a feed-forward approach for 360{\deg} novel view synthesis (NVS) of diverse real-world scenes, using only sparse observations. This setting is inherently ill-posed due to minimal overlap among input views and insufficient visual information provided, making it challenging for conventional methods to achieve high-quality results. Our MVSplat360 addresses this by effectively combining geometry-aware 3D reconstruction with temporally consistent video generation. Specifically, it refactors a feed-forward 3D Gaussian Splatting (3DGS) model to render features directly into the latent space of a pre-trained Stable Video Diffusion (SVD) model, where these features then act as pose and visual cues to guide the denoising process and produce photorealistic 3D-consistent views. Our model is end-to-end trainable and supports rendering arbitrary views with as few as 5 sparse input views. To evaluate MVSplat360's performance, we introduce a new benchmark using the challenging DL3DV-10K dataset, where MVSplat360 achieves superior visual quality compared to state-of-the-art methods on wide-sweeping or even 360{\deg} NVS tasks. Experiments on the existing benchmark RealEstate10K also confirm the effectiveness of our model. The video results are available on our project page: https://donydchen.github.io/mvsplat360.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

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

  1. TinySplat: Feedforward Approach for Generating Compact 3D Scene Representation

    cs.CV 2025-06 conditional novelty 7.0 of 10

    TinySplat compresses feedforward 3D Gaussian scenes by 105-199x on two-view benchmarks (about 50x on DL3DV) while keeping rendered quality close to the uncompressed model.

  2. iLRM: An Iterative Large 3D Reconstruction Model

    cs.CV 2025-07 conditional novelty 6.0 of 10

    iLRM reconstructs 3D Gaussian scenes from multiple photos through iterative refinement of viewpoint tokens, achieving higher quality and speed than prior feed-forward models.

  3. EarthCrafter: Scalable 3D Earth Generation via Dual-Sparse Latent Diffusion

    cs.CV 2025-07 conditional novelty 6.0 of 10

    EarthCrafter generates 600-meter-scale 3D Earth scenes using separate latent diffusion models for structure and texture, conditioned on semantics, images, or nothing.

  4. LongSplat: Online Generalizable 3D Gaussian Splatting from Long Sequence Images

    cs.CV 2025-07 reject novelty 6.0 of 10

    A feed-forward 3D Gaussian Splatting pipeline that incrementally fuses and compresses historical Gaussians using a 2D image-like representation.

  5. Sparse-View 3D Reconstruction: Recent Advances and Open Challenges

    cs.CV 2025-07 conditional novelty 3.0 of 10

    A comprehensive survey that organizes sparse-view 3D reconstruction methods into geometry-based, NeRF, 3DGS, and diffusion-based categories, with benchmarks and open challenges.

Pith tools