Research map / 3D vision / Gaussian splatting
MIGS: Multi-Identity Gaussian Splatting via Tensor Decomposition
Aggelina Chatziagapi, Grigorios Chrysos, Dimitris Samaras
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 |
|---|---|---|---|
| learn a single multi-identity CP-decomposed 3DGS representation instead of per-identity 3DGS-Avatar | MIGS · AIST++ Advanced Test | PSNR: 25.51 -> 26.54 | Improved |
| learn multi-identity representation from 30 identities on AIST++ Basic Test | MIGS · AIST++ Basic Test | PSNR: 28.89 -> 29.82 | Improved |
| multi-identity training versus single-identity 3DGS-Avatar on novel view synthesis | MIGS · ZJU-MoCap subject 387 | LPIPS*: 34.24 -> 35.33 | Mixed |
| use CP rank R=100 instead of R=10 (ablation) | MIGS · AIST++ Advanced Test | LPIPS* | Improved |
| increase rank from R=100 to R=200 with personalization (ablation) | MIGS · AIST++ Advanced Test | LPIPS* | No clear effect |
| increase number of training identities (ablation) | MIGS · AIST++ Advanced Test | LPIPS* | Improved |
| add personalization by fine-tuning color MLP only (ablation) | MIGS · AIST++ Advanced Test | visual quality | Improved |
| fine-tune all parameters instead of only color MLP during personalization (ablation) | MIGS · AIST++ novel poses | artifacts (qualitative) | Hurt |
| update only the identity's row of U2 versus the full U2 each iteration (ablation) | MIGS · training | visual quality | No clear effect |
| remove adaptive densification and pruning of Gaussians (ablation) | MIGS single identity · single identity | visual quality | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- 3DGS-Avatar — 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024) · better on 10 of 11 shared results
- HumanNeRF — HumanNeRF: Efficiently Generated Human Radiance Field From Sparse Inputs (CVPR 2022) · better on 3 of 3 shared results
- MonoHuman — MonoHuman: Animatable Human Neural Field From Monocular Video (CVPR 2023) · better on 2 of 2 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
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.89)
- Human Gaussian Splatting: Real-time Rendering of Animatable Avatars (CVPR 2024 · similarity 0.89)
- GauHuman: Articulated Gaussian Splatting from Monocular Human Videos (CVPR 2024 · similarity 0.89)
- Learning Interaction-aware 3D Gaussian Splatting for One-shot Hand Avatars (NeurIPS 2024 · similarity 0.88)
- HUGS: Human Gaussian Splats (CVPR 2024 · similarity 0.88)
- COLMAP-Free 3D Gaussian Splatting (CVPR 2024 · similarity 0.88)
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.
