Research map / 3D vision / Gaussian splatting
DcSplat: Dual-Constraint Human Gaussian Splatting with Latent Multi-View Consistency
Tengfei Xiao, Yue Wu, Zhigang Gao, Yongzhe Yuan, Can Qin, Hao Li, Mingyang 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 SMPL-based pseudo views (ablation) | DcSplat · THuman2.1 | PSNR: 23.873 -> 26.639 | Improved |
| add Depth-aware Appearance Injector (DAI) (ablation) | DcSplat · THuman2.1 | PSNR: 25.951 -> 26.639 | Improved |
| add Geometric Constraint Module (GCM) (ablation) | DcSplat · THuman2.1 | PSNR: 25.606 -> 26.639 | Improved |
| add GCM (ablation) | DcSplat · THuman2.1 | LPIPS: 0.0612 -> 0.0452 | Improved |
| use denser pseudo-view spacing (30 and 60 degrees vs 180) (ablation) | DcSplat · THuman2.1 | PSNR / inference time: inference time at 30 degrees approximately 8x that of 180 degrees | Mixed |
| use 90 degree pseudo-view spacing instead of denser ones (ablation) | DcSplat · THuman2.1 | quality vs efficiency | Improved |
| use DcSplat instead of RoGSplat | single-view human NVS · Real-World (cross-domain) | PSNR: 24.209 -> 25.753 | Improved |
| use DcSplat instead of RoGSplat | single-view human NVS · THuman2.1 (in-domain) | PSNR: 26.748 -> 26.639 | Hurt |
| use DcSplat instead of LHM | single-view human NVS · Real-World inference cost | parameters / time: 500 M -> 29.3 M, 2.01 s -> 0.07 s | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- RoGSplat — RoGSplat: Learning Robust Generalizable Human Gaussian Splatting from Sparse Multi-View Images (CVPR 2025) · better on 6 of 8 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
- HumanSplat: Generalizable Single-Image Human Gaussian Splatting with Structure Priors (NeurIPS 2024 · similarity 0.92)
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · similarity 0.90)
- RoGSplat: Learning Robust Generalizable Human Gaussian Splatting from Sparse Multi-View Images (CVPR 2025 · similarity 0.89)
- Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.89)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (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.
