Research map / 3D vision / Gaussian splatting
VoluMe - Authentic 3D Video Calls from Live Gaussian Splat Prediction
Martin de La Gorce, Charlie Hewitt, Tibor Takács, Robert Gerdisch, Zafiirah Hosenie, Givi Meishvili, Marek Kowalski, Thomas J. Cashman, Antonio Criminisi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add jitter stability loss L_j (ablation) | VoluMe · Ava256 250 subjects | Jitter: 0.005020 -> 0.004967 | Improved |
| add jitter stability loss L_j (ablation) | VoluMe · Ava256 250 subjects | LPIPS: 0.1581 -> 0.1497 | Improved |
| predict two Gaussians per pixel instead of one (ablation) | VoluMe · Ava-256 ablation | PSNR: 18.907 -> 19.069 | Improved |
| add ROI homography instead of simple cropping (ablation) | VoluMe · Ava-256 ablation | PSNR: 18.901 -> 19.069 | Improved |
| add direct colour sampling (ablation) | VoluMe · Ava-256 ablation | PSNR: 19.026 -> 19.069 | Improved |
| add optimizable layers (ablation) | VoluMe · Ava-256 ablation | PSNR: 19.005 -> 19.069 | Improved |
| combine all five features vs baseline (ablation) | VoluMe · Ava-256 ablation | PSNR: 18.698 -> 19.069 | Improved |
| combine all five features vs baseline (ablation) | VoluMe · Ava-256 ablation | Jitter: 0.004959 -> 0.004967 | No clear effect |
| use two Gaussians per pixel instead of one (ablation) | VoluMe · Ava-256 (60 epochs) | PSNR: 19.07 -> 19.15 | Improved |
| use three Gaussians per pixel instead of two (ablation) | VoluMe · Ava-256 (60 epochs) | PSNR: 19.15 -> 18.99 | Hurt |
| use four Gaussians per pixel instead of two (ablation) | VoluMe · Ava-256 (60 epochs) | Jitter: 0.005086 -> 0.005990 | Hurt |
| use VoluMe instead of Splatter Image | VoluMe · Cafca subset | LPIPS: 0.2244 -> 0.1868 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Splatter Image — Splatter Image: Ultra-Fast Single-View 3D Reconstruction (CVPR 2024) · better on 7 of 7 shared results
- FaceLift — FaceLift: Learning Generalizable Single Image 3D Face Reconstruction from Synthetic Heads (ICCV 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Motion Matters: Compact Gaussian Streaming for Free-Viewpoint Video Reconstruction (NeurIPS 2025 · similarity 0.85)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · similarity 0.84)
- GauHuman: Articulated Gaussian Splatting from Monocular Human Videos (CVPR 2024 · similarity 0.84)
- Event-3DGS: Event-based 3D Reconstruction Using 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.84)
- RePerformer: Immersive Human-centric Volumetric Videos from Playback to Photoreal Reperformance (CVPR 2025 · similarity 0.84)
- Human Gaussian Splatting: Real-time Rendering of Animatable Avatars (CVPR 2024 · similarity 0.84)
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.
