${M^2D}$NeRF: Multi-Modal Decomposition NeRF with 3D Feature Fields

• URL: http://arxiv.org/abs/2405.05010v1
• Date: Wed, 8 May 2024 12:25:21 GMT
• Title: ${M^2D}$NeRF: Multi-Modal Decomposition NeRF with 3D Feature Fields
• Authors: Ning Wang, Lefei Zhang, Angel X Chang,
• Abstract summary: We present a single model, Multi-Modal Decomposition NeRF ($M2D$NeRF), that is capable of both text-based and visual patch-based edits. Specifically, we use multi-modal feature distillation to integrate teacher features from pretrained visual and language models into 3D semantic feature volumes. Experiments on various real-world scenes show superior performance in 3D scene decomposition tasks compared to prior NeRF-based methods.
• Score: 33.168225243348786
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural fields (NeRF) have emerged as a promising approach for representing continuous 3D scenes. Nevertheless, the lack of semantic encoding in NeRFs poses a significant challenge for scene decomposition. To address this challenge, we present a single model, Multi-Modal Decomposition NeRF (${M^2D}$NeRF), that is capable of both text-based and visual patch-based edits. Specifically, we use multi-modal feature distillation to integrate teacher features from pretrained visual and language models into 3D semantic feature volumes, thereby facilitating consistent 3D editing. To enforce consistency between the visual and language features in our 3D feature volumes, we introduce a multi-modal similarity constraint. We also introduce a patch-based joint contrastive loss that helps to encourage object-regions to coalesce in the 3D feature space, resulting in more precise boundaries. Experiments on various real-world scenes show superior performance in 3D scene decomposition tasks compared to prior NeRF-based methods.

Related papers

• DistillNeRF: Perceiving 3D Scenes from Single-Glance Images by Distilling Neural Fields and Foundation Model Features [65.8738034806085]
  DistillNeRF is a self-supervised learning framework for understanding 3D environments in autonomous driving scenes. Our method is a generalizable feedforward model that predicts a rich neural scene representation from sparse, single-frame multi-view camera inputs.
  arXiv  Detail & Related papers  (2024-06-17T21:15:13Z)
• Denoising Diffusion via Image-Based Rendering [54.20828696348574]
  We introduce the first diffusion model able to perform fast, detailed reconstruction and generation of real-world 3D scenes. First, we introduce a new neural scene representation, IB-planes, that can efficiently and accurately represent large 3D scenes. Second, we propose a denoising-diffusion framework to learn a prior over this novel 3D scene representation, using only 2D images.
  arXiv  Detail & Related papers  (2024-02-05T19:00:45Z)
• Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields [54.482261428543985]
  Methods that use Neural Radiance fields are versatile for traditional tasks such as novel view synthesis. 3D Gaussian splatting has shown state-of-the-art performance on real-time radiance field rendering. We propose architectural and training changes to efficiently avert this problem.
  arXiv  Detail & Related papers  (2023-12-06T00:46:30Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [114.47216525866435]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• CompoNeRF: Text-guided Multi-object Compositional NeRF with Editable 3D Scene Layout [13.364394556439992]
  Text-to-3D form plays a crucial role in creating editable 3D scenes for AR/VR. Recent advances have shown promise in merging neural radiance fields (NeRFs) with pre-trained diffusion models for text-to-3D object generation. We propose a novel framework, dubbed CompoNeRF, by integrating an editable 3D scene layout with object-specific and scene-wide guidance mechanisms. Our framework achieves up to a textbf54% improvement by the multi-view CLIP score metric.
  arXiv  Detail & Related papers  (2023-03-24T07:37:09Z)
• NerfDiff: Single-image View Synthesis with NeRF-guided Distillation from 3D-aware Diffusion [107.67277084886929]
  Novel view synthesis from a single image requires inferring occluded regions of objects and scenes whilst simultaneously maintaining semantic and physical consistency with the input. We propose NerfDiff, which addresses this issue by distilling the knowledge of a 3D-aware conditional diffusion model (CDM) into NeRF through synthesizing and refining a set of virtual views at test time. We further propose a novel NeRF-guided distillation algorithm that simultaneously generates 3D consistent virtual views from the CDM samples, and finetunes the NeRF based on the improved virtual views.
  arXiv  Detail & Related papers  (2023-02-20T17:12:00Z)
• Neural Feature Fusion Fields: 3D Distillation of Self-Supervised 2D Image Representations [92.88108411154255]
  We present a method that improves dense 2D image feature extractors when the latter are applied to the analysis of multiple images reconstructible as a 3D scene. We show that our method not only enables semantic understanding in the context of scene-specific neural fields without the use of manual labels, but also consistently improves over the self-supervised 2D baselines.
  arXiv  Detail & Related papers  (2022-09-07T23:24:09Z)
• Decomposing NeRF for Editing via Feature Field Distillation [14.628761232614762]
  editing a scene represented by a NeRF is challenging as the underlying connectionist representations are not object-centric or compositional. In this work, we tackle the problem of semantic scene decomposition of NeRFs to enable query-based local editing. We propose to distill the knowledge of off-the-shelf, self-supervised 2D image feature extractors into a 3D feature field optimized in parallel to the radiance field.
  arXiv  Detail & Related papers  (2022-05-31T07:56:09Z)
• 3D-aware Image Synthesis via Learning Structural and Textural Representations [39.681030539374994]
  We propose VolumeGAN, for high-fidelity 3D-aware image synthesis, through explicitly learning a structural representation and a textural representation. Our approach achieves sufficiently higher image quality and better 3D control than the previous methods.
  arXiv  Detail & Related papers  (2021-12-20T18:59:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.