Research map / 3D vision / Face and shape
NaviNeRF: NeRF-based 3D Representation Disentanglement by Latent Semantic Navigation
Baao Xie, Bohan Li, Zequn Zhang, Junting Dong, Xin Jin, Jingyu Yang, Wenjun Zeng
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 |
|---|---|---|---|
| shift only the 9th-18th dimensions of w instead of all 18 (ablation) | NaviNeRF inner branch · FFHQ (qualitative) | disentanglement of fine attributes | Improved |
| use W+ space instead of W space for shifts (ablation) | NaviNeRF inner branch · FFHQ (qualitative) | fine-grained manipulation | Improved |
| add synergistic loss coupling the two branches (ablation) | NaviNeRF · FFHQ (qualitative) | 3D consistency | Improved |
| NaviNeRF versus StyleNeRF image quality | NaviNeRF · FFHQ | FID: 10.4 -> 13 | Hurt |
| NaviNeRF versus GIRAFFE and pi-GAN image quality | NaviNeRF · FFHQ | FID: 38 -> 13 (GIRAFFE), 87 -> 13 (pi-GAN) | Improved |
| add second (inner) branch | NaviNeRF · FFHQ training budget | training time (hrs) / parameters (M): 5.8 -> 7.5 hrs; 11.9 -> 12.2 M | Hurt |
| NaviNeRF versus pure NeRF-based methods in rendering | NaviNeRF · Rendering 256x256 | ms/image: 785 -> 97 (vs pi-GAN) | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- StyleNeRF — StyleNeRF: A Style-based 3D Aware Generator for High-resolution Image Synthesis (ICLR 2022) · better on 0 of 5 shared results
- GIRAFFE — GIRAFFE: Representing Scenes As Compositional Generative Neural Feature Fields (CVPR 2021) · better on 4 of 4 shared results
- pi-GAN — Pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis (CVPR 2021) · 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
- NaviNeRF++: Towards Interpretable 3D Reconstruction via Unsupervised Disentangled Representation Learning (TPAMI 2025 · similarity 0.92)
- Closed-Loop Unsupervised Representation Disentanglement with $\\beta$-VAE Distillation and Diffusion Probabilistic Feedback (ECCV 2024 · similarity 0.86)
- Decomposing NeRF for Editing via Feature Field Distillation (NeurIPS 2022 · similarity 0.86)
- Disentangled 3D Scene Generation with Layout Learning (ICML 2024 · similarity 0.85)
- NeRF Analogies: Example-Based Visual Attribute Transfer for NeRFs (CVPR 2024 · similarity 0.85)
- Chain of Semantics Programming in 3D Gaussian Splatting Representation for 3D Vision Grounding (CVPR 2025 · similarity 0.84)
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.
