mlsanityData to 7 Oct 2026

NaviNeRF: NeRF-based 3D Representation Disentanglement by Latent Semantic Navigation

arXiv 2304.11342 ↗First posted 22 Apr 2023cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

3D representation disentanglement aims to identify, decompose, and manipulate the underlying explanatory factors of 3D data, which helps AI fundamentally understand our 3D world. This task is currently under-explored and poses great challenges: (i) the 3D representations are complex and in general contains much more information than 2D image; (ii) many 3D representations are not well suited for gradient-based optimization, let alone disentanglement. To address these challenges, we use NeRF as a differentiable 3D representation, and introduce a self-supervised Navigation to identify interpretable semantic directions in the latent space. To our best knowledge, this novel method, dubbed NaviNeRF, is the first work to achieve fine-grained 3D disentanglement without any priors or supervisions. Specifically, NaviNeRF is built upon the generative NeRF pipeline, and equipped with an Outer Navigation Branch and an Inner Refinement Branch. They are complementary -- the outer navigation is to identify global-view semantic directions, and the inner refinement dedicates to fine-grained attributes. A synergistic loss is further devised to coordinate two branches. Extensive experiments demonstrate that NaviNeRF has a superior fine-grained 3D disentanglement ability than the previous 3D-aware models. Its performance is also comparable to editing-oriented models relying on semantic or geometry priors.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 50 of its 62 references that are in the map. First posted on arXiv 22 Apr 2023.

Builds on

50 of 62 references.

References · 50

  1. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  2. 2022Generative Adversarial Networkscited by 5,603
  3. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  5. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  6. 2020D-NeRF: Neural Radiance Fields for Dynamic Scenescited by 1,459
  7. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  8. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,230
  9. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
  10. 2021StyleCLIP: Text-Driven Manipulation of StyleGAN Imagerycited by 990
  11. 2020NeRF++: Analyzing and Improving Neural Radiance Fieldscited by 857
  12. 2020GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fieldscited by 784
38 more
  1. 2020Encoding in Style: a StyleGAN Encoder for Image-to-Image Translationcited by 727
  2. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  3. 2019Interpreting the Latent Space of GANs for Semantic Face Editingcited by 691
  4. 2020GRAF: Generative Radiance Fields for 3D-Aware Image Synthesiscited by 688
  5. 2021StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesiscited by 477
  6. 2020Closed-Form Factorization of Latent Semantics in GANscited by 407
  7. 2021CLIP-NeRF: Text-and-Image Driven Manipulation of Neural Radiance Fieldscited by 369
  8. 2021GAN Inversion: A Surveycited by 366
  9. 2016Generative Visual Manipulation on the Natural Image Manifoldcited by 348
  10. 2020Unsupervised Discovery of Interpretable Directions in the GAN Latent Spacecited by 288
  11. 2019On the "steerability" of generative adversarial networkscited by 247
  12. 2021HeadNeRF: A Real-time NeRF-based Parametric Head Modelcited by 228
  13. 2022NeRF-Editing: Geometry Editing of Neural Radiance Fieldscited by 216
  14. 2019GANalyze: Toward Visual Definitions of Cognitive Image Propertiescited by 193
  15. 2021CodeNeRF: Disentangled Neural Radiance Fields for Object Categoriescited by 183
  16. 2021NeRF--: Neural Radiance Fields Without Known Camera Parameterscited by 155
  17. 2021GNeRF: GAN-based Neural Radiance Field without Posed Cameracited by 153
  18. 2015A Large-Scale Car Dataset for Fine-Grained Categorization and Verificationcited by 129
  19. 2021FENeRF: Face Editing in Neural Radiance Fieldscited by 112
  20. 2021NeRF-VAE: A Geometry Aware 3D Scene Generative Modelcited by 111
  21. 2020Controlling generative models with continuous factors of variationscited by 93
  22. 2022SDF-StyleGAN: Implicit SDF-Based StyleGAN for 3D Shape Generationcited by 92
  23. 2020The Hessian Penalty: A Weak Prior for Unsupervised Disentanglementcited by 90
  24. 2022DFA-NeRF: Personalized Talking Head Generation via Disentangled Face Attributes Neural Renderingcited by 80
  25. 2020Transforming and Projecting Images into Class-conditional Generative Networkscited by 76
  26. 2022Pix2NeRF: Unsupervised Conditional π-GAN for Single Image to Neural Radiance Fields Translationcited by 73
  27. 2021CoNeRF: Controllable Neural Radiance Fieldscited by 65
  28. 2021Generative Occupancy Fields for 3D Surface-Aware Image Synthesiscited by 51
  29. 2020Navigating the GAN Parameter Space for Semantic Image Editingcited by 49
  30. 2022Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learningcited by 46
  31. 2022Controllable 3D Face Synthesis with Conditional Generative Occupancy Fieldscited by 37
  32. 2021The Geometry of Deep Generative Image Models and its Applicationscited by 34
  33. 2021Learning Disentangled Representation by Exploiting Pretrained Generative Models: A Contrastive Learning Viewcited by 34
  34. 2021CG-NeRF: Conditional Generative Neural Radiance Fieldscited by 18
  35. 2021Disentangled Representations from Non-Disentangled Modelscited by 13
  36. 2022AR-NeRF: Unsupervised Learning of Depth and Defocus Effects from Natural Images with Aperture Rendering Neural Radiance Fieldscited by 12
  37. 2022Generative Deformable Radiance Fields for Disentangled Image Synthesis of Topology-Varying Objectscited by 8
  38. 2022Structure-Aware NeRF without Posed Camera via Epipolar Constraintcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 3; 2025: 2; 2026: 1.

8 papers.

  1. 2024Neural Radiance Field-based Visual Rendering: A Comprehensive Reviewcited by 12
  2. 2023Targeted Adversarial Attacks on Generalizable Neural Radiance Fieldscited by 10
  3. 2024Closed-Loop Unsupervised Representation Disentanglement with β-VAE Distillation and Diffusion Probabilistic Feedbackcited by 9
  4. 2024Graph-based Unsupervised Disentangled Representation Learning via Multimodal Large Language Modelscited by 9
  5. 2025Disentangled World Models: Learning to Transfer Semantic Knowledge from Distracting Videos for Reinforcement Learningcited by 6
  6. 2025Interpretable Single-View 3D Gaussian Splatting using Unsupervised Hierarchical Disentangled Representation Learningcited by 1
  7. 2023BID-NeRF: RGB-D image pose estimation with inverted Neural Radiance Fieldscited by 0
  8. 2026First Shape, Then Meaning: Efficient Geometry and Semantics Learning for Indoor Reconstructioncited by 0