Research map / 3D vision / 3D generation
DSAR: Dual-Stream Autoregressive Modeling of Temporal Cloth Dynamics for Photorealistic Animatable Avatars
Xiong Haozhong, Yao Yu, Yu Zhou, Sidan DU
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 adaptive temporal regularization (ablation) | DSAR · Near-distribution test | PSNR: 29.3130 -> 31.7153 | Improved |
| add adaptive temporal regularization (ablation) | DSAR · Far-distribution test | PSNR: 28.8778 -> 30.6720 | Improved |
| add geometric stream (previous deformation feedback) (ablation) | DSAR · Far-distribution test | PSNR: 24.2694 -> 30.6720 | Improved |
| add geometric stream (previous deformation feedback) (ablation) | DSAR · Near-distribution test | PSNR: 28.9622 -> 31.7153 | Improved |
| add state stream (memory bank) (ablation) | DSAR · Far-distribution test | PSNR: 28.2286 -> 30.6720 | Improved |
| add state stream (memory bank) (ablation) | DSAR · Near-distribution test | PSNR: 30.2593 -> 31.7153 | Improved |
| add velocity map (ablation) | DSAR · Far-distribution test | FVD: 1.87 -> 0.23 | Improved |
| use causal attention instead of bidirectional attention (ablation) | DSAR · Far-distribution test | FVD: 5.31 -> 0.23 | Improved |
| use memory bank state stream instead of direct cross-attention on previous frame (ablation) | DSAR · Far-distribution test | FVD: 2.14 -> 0.23 | Improved |
| add motion embedding (ablation) | DSAR · Far-distribution test | FVD: 1.94 -> 0.23 | Improved |
| use state stream instead of long skeletal pose history (ablation) | DSAR · Far-distribution test | FVD: 6.86 -> 0.23 | Improved |
| condition on deformations of previous 3 frames instead of one (ablation) | DSAR · Far-distribution test | PSNR: 30.6720 -> 30.7161 | No clear effect |
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.
- GaussianAvatar — GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (CVPR 2024) · better on 3 of 4 shared results
- HumanNeRF — HumanNeRF: Efficiently Generated Human Radiance Field From Sparse Inputs (CVPR 2022) · 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
- MPMAvatar: Learning 3D Gaussian Avatars with Accurate and Robust Physics-Based Dynamics (NeurIPS 2025 · similarity 0.87)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.87)
- IntrinsicAvatar: Physically Based Inverse Rendering of Dynamic Humans from Monocular Videos via Explicit Ray Tracing (CVPR 2024 · similarity 0.87)
- Pursuing Temporal-Consistent Video Virtual Try-On via Dynamic Pose Interaction (CVPR 2025 · similarity 0.87)
- Human Gaussian Splatting: Real-time Rendering of Animatable Avatars (CVPR 2024 · similarity 0.86)
- ASH: Animatable Gaussian Splats for Efficient and Photoreal Human Rendering (CVPR 2024 · similarity 0.86)
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.
