Research map / 3D vision / Human pose estimation
LVSPM: Long Sequence View Synthesis and Pose Estimation Model
Xi Chen, Yachi Zhang, Linghao Chen, Minghua Liu, Hao Su, Zexiang Xu, Xiaoshuai Zhang
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 NVS (RGB) supervision to pose training (ablation) | LVSPM · ASE ablation pose (32 views) | AUC3: 71.9 vs 50.9 AUC3 (w/o RGB) | Improved |
| add large-scale synthetic (ASE) pretraining (ablation) | LVSPM · DL3DV-10K (32 views) | pose and NVS | Improved |
| use full chunk size 8224 with one update instead of half chunk 4112 with two updates (ablation) | LVSPM · ASE ablation (32 views) | pose and NVS: 8224 vs 4112 tokens | Improved |
| use 24 TTT layers instead of 12 (ablation) | LVSPM · ASE ablation (32 views) | pose and NVS: 24 vs 12 layers | Improved |
| use LVSPM instead of VGGT | LVSPM · RE10k pose (16 views) | AUC3: 14.0 -> 40.0 | Improved |
| use LVSPM instead of VGGT | LVSPM · RE10k pose (128 views) | AUC3: 20.6 -> 61.3 | Improved |
| use LVSPM instead of VGGT | LVSPM · Co3Dv2 pose (128 views) | AUC3: 62.1 -> 68.8 | Improved |
| use LVSPM instead of DA3 | LVSPM · Co3Dv2 pose (16 views) | AUC3: 47.2 -> 39.2 | Hurt |
| use LVSPM instead of VGGT | LVSPM · DL3DV-10K pose (16 views) | AUC3: 74.7 -> 88.4 | Improved |
| use LVSPM instead of VGGT | LVSPM · DL3DV-10K pose (256 views) | AUC3: 83.6 -> 90.1 | Improved |
| use LVSPM instead of AnySplat | LVSPM · DL3DV-10K NVS (256 views) | PSNR: 18.55 -> 22.09 | Improved |
| use LVSPM instead of pose-dependent DepthSplat | LVSPM · DL3DV-10K NVS (16 views) | PSNR: 24.80 -> 25.77 | 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.
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 7 of 7 shared results
- DepthSplat — DepthSplat: Connecting Gaussian Splatting and Depth (CVPR 2025) · 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
- LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias (ICLR 2025 · similarity 0.89)
- NVComposer: Boosting Generative Novel View Synthesis with Multiple Sparse and Unposed Images (CVPR 2025 · similarity 0.85)
- MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views (NeurIPS 2024 · similarity 0.85)
- PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization (CVPR 2024 · similarity 0.85)
- MVGenMaster: Scaling Multi-View Generation from Any Image via 3D Priors Enhanced Diffusion Model (CVPR 2025 · similarity 0.85)
- NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis (NeurIPS 2026 · similarity 0.85)
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.
