Normal-guided Detail-Preserving Neural Implicit Functions for High-Fidelity 3D Surface Reconstruction

• URL: http://arxiv.org/abs/2406.04861v1
• Date: Fri, 7 Jun 2024 11:48:47 GMT
• Title: Normal-guided Detail-Preserving Neural Implicit Functions for High-Fidelity 3D Surface Reconstruction
• Authors: Aarya Patel, Hamid Laga, Ojaswa Sharma,
• Abstract summary: Current methods for learning neural implicit representations from RGB or RGBD images produce 3D surfaces with missing parts and details. This paper demonstrates that training neural representations with first-order differential properties, i.e. surface normals, leads to highly accurate 3D surface reconstruction.
• Score: 6.4279213810512665
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Neural implicit representations have emerged as a powerful paradigm for 3D reconstruction. However, despite their success, existing methods fail to capture fine geometric details and thin structures, especially in scenarios where only sparse RGB views of the objects of interest are available. We hypothesize that current methods for learning neural implicit representations from RGB or RGBD images produce 3D surfaces with missing parts and details because they only rely on 0-order differential properties, i.e. the 3D surface points and their projections, as supervisory signals. Such properties, however, do not capture the local 3D geometry around the points and also ignore the interactions between points. This paper demonstrates that training neural representations with first-order differential properties, i.e. surface normals, leads to highly accurate 3D surface reconstruction even in situations where only as few as two RGB (front and back) images are available. Given multiview RGB images of an object of interest, we first compute the approximate surface normals in the image space using the gradient of the depth maps produced using an off-the-shelf monocular depth estimator such as Depth Anything model. An implicit surface regressor is then trained using a loss function that enforces the first-order differential properties of the regressed surface to match those estimated from Depth Anything. Our extensive experiments on a wide range of real and synthetic datasets show that the proposed method achieves an unprecedented level of reconstruction accuracy even when using as few as two RGB views. The detailed ablation study also demonstrates that normal-based supervision plays a key role in this significant improvement in performance, enabling the 3D reconstruction of intricate geometric details and thin structures that were previously challenging to capture.

Related papers

• ANIM: Accurate Neural Implicit Model for Human Reconstruction from a single RGB-D image [40.03212588672639]
  ANIM is a novel method that reconstructs arbitrary 3D human shapes from single-view RGB-D images with an unprecedented level of accuracy. Our model learns geometric details from both pixel-aligned and voxel-aligned features to leverage depth information. Experiments demonstrate that ANIM outperforms state-of-the-art works that use RGB, surface normals, point cloud or RGB-D data as input.
  arXiv  Detail & Related papers  (2024-03-15T14:45:38Z)
• MonoNPHM: Dynamic Head Reconstruction from Monocular Videos [47.504979561265536]
  We present Monocular Neural Parametric Head Models (MonoNPHM) for dynamic 3D head reconstructions from monocular RGB videos. We constrain predicted color values to be correlated with the underlying geometry such that gradients from RGB effectively influence latent geometry codes during inverse rendering.
  arXiv  Detail & Related papers  (2023-12-11T17:55:05Z)
• NeuSG: Neural Implicit Surface Reconstruction with 3D Gaussian Splatting Guidance [59.08521048003009]
  We propose a neural implicit surface reconstruction pipeline with guidance from 3D Gaussian Splatting to recover highly detailed surfaces. The advantage of 3D Gaussian Splatting is that it can generate dense point clouds with detailed structure. We introduce a scale regularizer to pull the centers close to the surface by enforcing the 3D Gaussians to be extremely thin.
  arXiv  Detail & Related papers  (2023-12-01T07:04:47Z)
• Flattening-Net: Deep Regular 2D Representation for 3D Point Cloud Analysis [66.49788145564004]
  We present an unsupervised deep neural architecture called Flattening-Net to represent irregular 3D point clouds of arbitrary geometry and topology. Our methods perform favorably against the current state-of-the-art competitors.
  arXiv  Detail & Related papers  (2022-12-17T15:05:25Z)
• Recovering Fine Details for Neural Implicit Surface Reconstruction [3.9702081347126943]
  We present D-NeuS, a volume rendering neural implicit surface reconstruction method capable to recover fine geometry details. We impose multi-view feature consistency on the surface points, derived by interpolating SDF zero-crossings from sampled points along rays. Our method reconstructs high-accuracy surfaces with details, and outperforms the state of the art.
  arXiv  Detail & Related papers  (2022-11-21T10:06:09Z)
• GAN2X: Non-Lambertian Inverse Rendering of Image GANs [85.76426471872855]
  We present GAN2X, a new method for unsupervised inverse rendering that only uses unpaired images for training. Unlike previous Shape-from-GAN approaches that mainly focus on 3D shapes, we take the first attempt to also recover non-Lambertian material properties by exploiting the pseudo paired data generated by a GAN. Experiments demonstrate that GAN2X can accurately decompose 2D images to 3D shape, albedo, and specular properties for different object categories, and achieves the state-of-the-art performance for unsupervised single-view 3D face reconstruction.
  arXiv  Detail & Related papers  (2022-06-18T16:58:49Z)
• H3D-Net: Few-Shot High-Fidelity 3D Head Reconstruction [27.66008315400462]
  Recent learning approaches that implicitly represent surface geometry have shown impressive results in the problem of multi-view 3D reconstruction. We tackle these limitations for the specific problem of few-shot full 3D head reconstruction. We learn a shape model of 3D heads from thousands of incomplete raw scans using implicit representations.
  arXiv  Detail & Related papers  (2021-07-26T23:04:18Z)
• Deep Implicit Surface Point Prediction Networks [49.286550880464866]
  Deep neural representations of 3D shapes as implicit functions have been shown to produce high fidelity models. This paper presents a novel approach that models such surfaces using a new class of implicit representations called the closest surface-point (CSP) representation.
  arXiv  Detail & Related papers  (2021-06-10T14:31:54Z)
• Pix2Surf: Learning Parametric 3D Surface Models of Objects from Images [64.53227129573293]
  We investigate the problem of learning to generate 3D parametric surface representations for novel object instances, as seen from one or more views. We design neural networks capable of generating high-quality parametric 3D surfaces which are consistent between views. Our method is supervised and trained on a public dataset of shapes from common object categories.
  arXiv  Detail & Related papers  (2020-08-18T06:33:40Z)
• Geodesic-HOF: 3D Reconstruction Without Cutting Corners [42.4960665928525]
  Single-view 3D object reconstruction is a challenging fundamental problem in computer vision. We learn an image-conditioned mapping function from a canonical sampling domain to a high dimensional space. We find that this learned geodesic embedding space provides useful information for applications such as unsupervised object decomposition.
  arXiv  Detail & Related papers  (2020-06-14T18:59:06Z)
• Implicit Functions in Feature Space for 3D Shape Reconstruction and Completion [53.885984328273686]
  Implicit Feature Networks (IF-Nets) deliver continuous outputs, can handle multiple topologies, and complete shapes for missing or sparse input data. IF-Nets clearly outperform prior work in 3D object reconstruction in ShapeNet, and obtain significantly more accurate 3D human reconstructions.
  arXiv  Detail & Related papers  (2020-03-03T11:14:29Z)

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.