Decomposing NeRF for Editing via Feature Field Distillation
- URL: http://arxiv.org/abs/2205.15585v1
- Date: Tue, 31 May 2022 07:56:09 GMT
- Title: Decomposing NeRF for Editing via Feature Field Distillation
- Authors: Sosuke Kobayashi, Eiichi Matsumoto, Vincent Sitzmann
- Abstract summary: 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.
- Score: 14.628761232614762
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Emerging neural radiance fields (NeRF) are a promising scene representation
for computer graphics, enabling high-quality 3D reconstruction and novel view
synthesis from image observations. However, editing a scene represented by a
NeRF is challenging, as the underlying connectionist representations such as
MLPs or voxel grids are not object-centric or compositional. In particular, it
has been difficult to selectively edit specific regions or objects. In this
work, we tackle the problem of semantic scene decomposition of NeRFs to enable
query-based local editing of the represented 3D scenes. We propose to distill
the knowledge of off-the-shelf, self-supervised 2D image feature extractors
such as CLIP-LSeg or DINO into a 3D feature field optimized in parallel to the
radiance field. Given a user-specified query of various modalities such as
text, an image patch, or a point-and-click selection, 3D feature fields
semantically decompose 3D space without the need for re-training and enable us
to semantically select and edit regions in the radiance field. Our experiments
validate that the distilled feature fields (DFFs) can transfer recent progress
in 2D vision and language foundation models to 3D scene representations,
enabling convincing 3D segmentation and selective editing of emerging neural
graphics representations.
Related papers
- 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) - SERF: Fine-Grained Interactive 3D Segmentation and Editing with Radiance Fields [92.14328581392633]
We introduce a novel fine-grained interactive 3D segmentation and editing algorithm with radiance fields, which we refer to as SERF.
Our method entails creating a neural mesh representation by integrating multi-view algorithms with pre-trained 2D models.
Building upon this representation, we introduce a novel surface rendering technique that preserves local information and is robust to deformation.
arXiv Detail & Related papers (2023-12-26T02:50:42Z) - Learning Naturally Aggregated Appearance for Efficient 3D Editing [94.47518916521065]
We propose to replace the color field with an explicit 2D appearance aggregation, also called canonical image.
To avoid the distortion effect and facilitate convenient editing, we complement the canonical image with a projection field that maps 3D points onto 2D pixels for texture lookup.
Our representation, dubbed AGAP, well supports various ways of 3D editing (e.g., stylization, interactive drawing, and content extraction) with no need of re-optimization.
arXiv Detail & Related papers (2023-12-11T18:59:31Z) - Blended-NeRF: Zero-Shot Object Generation and Blending in Existing
Neural Radiance Fields [26.85599376826124]
We present Blended-NeRF, a framework for editing a specific region of interest in an existing NeRF scene.
We allow local editing by localizing a 3D ROI box in the input scene, and blend the content synthesized inside the ROI with the existing scene.
We show our framework for several 3D editing applications, including adding new objects to a scene, removing/altering existing objects, and texture conversion.
arXiv Detail & Related papers (2023-06-22T09:34:55Z) - Local 3D Editing via 3D Distillation of CLIP Knowledge [26.429032648560018]
3D content manipulation is an important computer vision task with many real-world applications.
Recent proposed 3D GANs can generate diverse photorealistic 3D-aware contents using Neural Radiance fields (NeRF)
We propose Local Editing NeRF (LENeRF), which only requires text inputs for fine-grained and localized manipulation.
arXiv Detail & Related papers (2023-06-21T21:09:45Z) - SINE: Semantic-driven Image-based NeRF Editing with Prior-guided Editing
Field [37.8162035179377]
We present a novel semantic-driven NeRF editing approach, which enables users to edit a neural radiance field with a single image.
To achieve this goal, we propose a prior-guided editing field to encode fine-grained geometric and texture editing in 3D space.
Our method achieves photo-realistic 3D editing using only a single edited image, pushing the bound of semantic-driven editing in 3D real-world scenes.
arXiv Detail & Related papers (2023-03-23T13:58:11Z) - Semantic 3D-aware Portrait Synthesis and Manipulation Based on
Compositional Neural Radiance Field [55.431697263581626]
We propose a Compositional Neural Radiance Field (CNeRF) for semantic 3D-aware portrait synthesis and manipulation.
CNeRF divides the image by semantic regions and learns an independent neural radiance field for each region, and finally fuses them and renders the complete image.
Compared to state-of-the-art 3D-aware GAN methods, our approach enables fine-grained semantic region manipulation, while maintaining high-quality 3D-consistent synthesis.
arXiv Detail & Related papers (2023-02-03T07:17:46Z) - PaletteNeRF: Palette-based Appearance Editing of Neural Radiance Fields [60.66412075837952]
We present PaletteNeRF, a novel method for appearance editing of neural radiance fields (NeRF) based on 3D color decomposition.
Our method decomposes the appearance of each 3D point into a linear combination of palette-based bases.
We extend our framework with compressed semantic features for semantic-aware appearance editing.
arXiv Detail & Related papers (2022-12-21T00:20:01Z) - 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) - Editing Conditional Radiance Fields [40.685602081728554]
A neural radiance field (NeRF) is a scene model supporting high-quality view synthesis, optimized per scene.
In this paper, we explore enabling user editing of a category-level NeRF trained on a shape category.
We introduce a method for propagating coarse 2D user scribbles to the 3D space, to modify the color or shape of a local region.
arXiv Detail & Related papers (2021-05-13T17:59:48Z)
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.