Research map / 3D vision / Face and shape
HiFi-Mesh: High-Fidelity Efficient 3D Mesh Generation via Compact Autoregressive Dependence
Yanfeng Li, Tao Tan, Qinquan Gao, Zhiwen Cao, Xiaohong Liu, Yue Sun
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 hierarchical latent space representations as autoregressive dependencies (ablation) | HiFi-Mesh · Objaverse mesh generation (ablation) | Chamfer distance (x10^-1): 1.291 -> 0.157 | Improved |
| add AdaGraph parallel inference strategy (ablation) | HiFi-Mesh · Objaverse mesh generation (ablation) | Inference speed (tokens/s): 121.7 -> 302.1 | Improved |
| HiFi-Mesh vs best baseline on normal consistency | HiFi-Mesh · Objaverse mesh generation | Normal consistency (x10^-1): 2.001 (TreeMeshGPT) vs 3.621 | Hurt |
| use latent autoregressive dependencies instead of long token history | HiFi-Mesh · Objaverse mesh generation | Maximum generatable length (tokens): 49,406 -> 300,000 | Improved |
| HiFi-Mesh vs EdgeRunner at short sequence lengths (under 5K) | HiFi-Mesh · memory and compute vs maximum sequence length | memory and TFLOPS | Hurt |
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.
- MeshAnythingV2 — MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025) · better on 7 of 7 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
- ARMesh: Autoregressive Mesh Generation via Next-Level-of-Detail Prediction (NeurIPS 2025 · similarity 0.90)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (CVPR 2024 · similarity 0.88)
- TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing (CVPR 2025 · similarity 0.87)
- Mesh BDF: Barycentric Dominance Field for 3D Native Mesh Generation (NeurIPS 2026 · similarity 0.87)
- EdgeRunner: Auto-regressive Auto-encoder for Artistic Mesh Generation (ICLR 2025 · similarity 0.87)
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · 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.
