Differentiable Point-Based Radiance Fields for Efficient View Synthesis

• URL: http://arxiv.org/abs/2205.14330v4
• Date: Wed, 5 Jul 2023 15:17:19 GMT
• Title: Differentiable Point-Based Radiance Fields for Efficient View Synthesis
• Authors: Qiang Zhang, Seung-Hwan Baek, Szymon Rusinkiewicz, Felix Heide
• Abstract summary: We propose a differentiable rendering algorithm for efficient novel view synthesis. Our method is up to 300x faster than NeRF in both training and inference. For dynamic scenes, our method trains two orders of magnitude faster than STNeRF and renders at near interactive rate.
• Score: 57.56579501055479
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose a differentiable rendering algorithm for efficient novel view synthesis. By departing from volume-based representations in favor of a learned point representation, we improve on existing methods more than an order of magnitude in memory and runtime, both in training and inference. The method begins with a uniformly-sampled random point cloud and learns per-point position and view-dependent appearance, using a differentiable splat-based renderer to evolve the model to match a set of input images. Our method is up to 300x faster than NeRF in both training and inference, with only a marginal sacrifice in quality, while using less than 10~MB of memory for a static scene. For dynamic scenes, our method trains two orders of magnitude faster than STNeRF and renders at near interactive rate, while maintaining high image quality and temporal coherence even without imposing any temporal-coherency regularizers.

Related papers

• Splatfacto-W: A Nerfstudio Implementation of Gaussian Splatting for Unconstrained Photo Collections [25.154665328053333]
  We introduce Splatfacto-W, an in-trivial approach that integrates per-Gaussian neural color features and per-image appearance embeddings into an rendering process. Our method improves the Peak Signal-to-Noise Ratio (PSNR) by an average of 5.3 dB compared to 3DGS, enhances training speed by 150 times compared to NeRF-based methods, and achieves a similar rendering speed to 3DGS.
  arXiv  Detail & Related papers  (2024-07-17T04:02:54Z)
• D-NPC: Dynamic Neural Point Clouds for Non-Rigid View Synthesis from Monocular Video [53.83936023443193]
  This paper contributes to the field by introducing a new synthesis method for dynamic novel view from monocular video, such as smartphone captures. Our approach represents the as a $textitdynamic neural point cloud$, an implicit time-conditioned point cloud that encodes local geometry and appearance in separate hash-encoded neural feature grids.
  arXiv  Detail & Related papers  (2024-06-14T14:35:44Z)
• INPC: Implicit Neural Point Clouds for Radiance Field Rendering [5.64500060725726]
  We introduce a new approach for reconstruction and novel-view synthesis of real-world scenes. We propose a hybrid scene representation, which implicitly encodes a point cloud in a continuous octree-based probability field and a multi-resolution hash grid. Our method achieves fast inference at interactive frame rates, and can extract explicit point clouds to further enhance performance.
  arXiv  Detail & Related papers  (2024-03-25T15:26:32Z)
• Temporal Interpolation Is All You Need for Dynamic Neural Radiance Fields [4.863916681385349]
  We propose a method to train neural fields of dynamic scenes based on temporal vectors of feature. In the neural representation, we extract from space-time inputs via multiple neural network modules and interpolate them based on time frames. In the grid representation, space-time features are learned via four-dimensional hash grids, which remarkably reduces training time.
  arXiv  Detail & Related papers  (2023-02-18T12:01:23Z)
• Multi-Frequency-Aware Patch Adversarial Learning for Neural Point Cloud Rendering [7.522462414919854]
  We present a neural point cloud rendering pipeline through a novel multi-frequency-aware patch adversarial learning framework. The proposed approach aims to improve the rendering realness by minimizing the spectrum discrepancy between real and synthesized images. Our method produces state-of-the-art results for neural point cloud rendering by a significant margin.
  arXiv  Detail & Related papers  (2022-10-07T16:54:15Z)
• Neural Deformable Voxel Grid for Fast Optimization of Dynamic View Synthesis [63.25919018001152]
  We propose a fast deformable radiance field method to handle dynamic scenes. Our method achieves comparable performance to D-NeRF using only 20 minutes for training.
  arXiv  Detail & Related papers  (2022-06-15T17:49:08Z)
• Fast Dynamic Radiance Fields with Time-Aware Neural Voxels [106.69049089979433]
  We propose a radiance field framework by representing scenes with time-aware voxel features, named as TiNeuVox. Our framework accelerates the optimization of dynamic radiance fields while maintaining high rendering quality. Our TiNeuVox completes training with only 8 minutes and 8-MB storage cost while showing similar or even better rendering performance than previous dynamic NeRF methods.
  arXiv  Detail & Related papers  (2022-05-30T17:47:31Z)
• View Synthesis with Sculpted Neural Points [64.40344086212279]
  Implicit neural representations have achieved impressive visual quality but have drawbacks in computational efficiency. We propose a new approach that performs view synthesis using point clouds. It is the first point-based method to achieve better visual quality than NeRF while being more than 100x faster in rendering speed.
  arXiv  Detail & Related papers  (2022-05-12T03:54:35Z)
• Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition [24.70946979449572]
  We develop a learning-based CNN+LSTM architecture, trainable via backpropagation through time, for viewpoint- and appearance-invariant place recognition. Our model outperforms 15 classical methods while setting new state-of-the-art performance standards. In addition, we show that SPL can be up to 70x faster to deploy than classical methods on a 729 km route.
  arXiv  Detail & Related papers  (2021-03-02T22:57:43Z)
• Displacement-Invariant Cost Computation for Efficient Stereo Matching [122.94051630000934]
  Deep learning methods have dominated stereo matching leaderboards by yielding unprecedented disparity accuracy. But their inference time is typically slow, on the order of seconds for a pair of 540p images. We propose a emphdisplacement-invariant cost module to compute the matching costs without needing a 4D feature volume.
  arXiv  Detail & Related papers  (2020-12-01T23:58:16Z)

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.