Research map / 3D vision / Gaussian splatting
StereoSplat: Metric-Scale Novel View Synthesis via Stereo-Grounded Gaussian Splatting
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes StereoSplat, a feed-forward stereo-video 3D Gaussian Splatting architecture that yields metric-scale novel-view synthesis by using disparity and features from a foundation stereo backbone.
Problem
Novel-view synthesis from stereo videos needs reliable, metric-scale 3D reconstruction, whereas monocular feed-forward methods ignore the binocular structure of stereo pairs.
Approach
A native stereo architecture that extracts dense multi-scale latent features and disparity from a foundation stereo backbone and feeds them into a dedicated Gaussian head, so Gaussians inherit metric scale from stereo geometry rather than monocular depth.
Experimental setup
Baselines: monocular feed-forward 3DGS methods
Builds on: 3D Gaussian Splatting
Closest papers by meaning
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · similarity 0.92)
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.90)
- MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views (NeurIPS 2024 · similarity 0.90)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (CVPR 2025 · similarity 0.90)
- Binocular-Guided 3D Gaussian Splatting with View Consistency for Sparse View Synthesis (NeurIPS 2024 · similarity 0.90)
- SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis (CVPR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.