End-to-end View Synthesis via NeRF Attention

• URL: http://arxiv.org/abs/2207.14741v2
• Date: Mon, 1 Aug 2022 03:53:27 GMT
• Title: End-to-end View Synthesis via NeRF Attention
• Authors: Zelin Zhao, Jiaya Jia
• Abstract summary: We present a simple seq2seq formulation for view synthesis where we take a set of ray points as input and output colors corresponding to the rays. Inspired by the neural radiance field (NeRF), we propose the NeRF attention (NeRFA) to address the above problems. NeRFA demonstrates superior performance over NeRF and NerFormer on four datasets.
• Score: 71.06080186332524
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In this paper, we present a simple seq2seq formulation for view synthesis where we take a set of ray points as input and output colors corresponding to the rays. Directly applying a standard transformer on this seq2seq formulation has two limitations. First, the standard attention cannot successfully fit the volumetric rendering procedure, and therefore high-frequency components are missing in the synthesized views. Second, applying global attention to all rays and pixels is extremely inefficient. Inspired by the neural radiance field (NeRF), we propose the NeRF attention (NeRFA) to address the above problems. On the one hand, NeRFA considers the volumetric rendering equation as a soft feature modulation procedure. In this way, the feature modulation enhances the transformers with the NeRF-like inductive bias. On the other hand, NeRFA performs multi-stage attention to reduce the computational overhead. Furthermore, the NeRFA model adopts the ray and pixel transformers to learn the interactions between rays and pixels. NeRFA demonstrates superior performance over NeRF and NerFormer on four datasets: DeepVoxels, Blender, LLFF, and CO3D. Besides, NeRFA establishes a new state-of-the-art under two settings: the single-scene view synthesis and the category-centric novel view synthesis. The code will be made publicly available.

Related papers

• NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning [63.39461847093663]
  We propose NeRF-VPT, an innovative method for novel view synthesis to address these challenges. Our proposed NeRF-VPT employs a cascading view prompt tuning paradigm, wherein RGB information gained from preceding rendering outcomes serves as instructive visual prompts for subsequent rendering stages. NeRF-VPT only requires sampling RGB data from previous stage renderings as priors at each training stage, without relying on extra guidance or complex techniques.
  arXiv  Detail & Related papers  (2024-03-02T22:08:10Z)
• Re-Nerfing: Improving Novel View Synthesis through Novel View Synthesis [80.3686833921072]
  Recent neural rendering and reconstruction techniques, such as NeRFs or Gaussian Splatting, have shown remarkable novel view synthesis capabilities. With fewer images available, these methods start to fail since they can no longer correctly triangulate the underlying 3D geometry. We propose Re-Nerfing, a simple and general add-on approach that leverages novel view synthesis itself to tackle this problem.
  arXiv  Detail & Related papers  (2023-12-04T18:56:08Z)
• Learning Neural Duplex Radiance Fields for Real-Time View Synthesis [33.54507228895688]
  We propose a novel approach to distill and bake NeRFs into highly efficient mesh-based neural representations. We demonstrate the effectiveness and superiority of our approach via extensive experiments on a range of standard datasets.
  arXiv  Detail & Related papers  (2023-04-20T17:59:52Z)
• Multiscale Tensor Decomposition and Rendering Equation Encoding for View Synthesis [7.680742911100444]
  We propose a novel approach dubbed the neural radiance feature field (NRFF) NRFF improves state-of-the-art rendering results by over 1 dB in PSNR on both the NeRF and NSVF datasets.
  arXiv  Detail & Related papers  (2023-03-07T11:21:50Z)
• Is Attention All NeRF Needs? [103.51023982774599]
  Generalizable NeRF Transformer (GNT) is a pure, unified transformer-based architecture that efficiently reconstructs Neural Radiance Fields (NeRFs) on the fly from source views. GNT achieves generalizable neural scene representation and rendering, by encapsulating two transformer-based stages.
  arXiv  Detail & Related papers  (2022-07-27T05:09:54Z)
• Generalizable Patch-Based Neural Rendering [46.41746536545268]
  We propose a new paradigm for learning models that can synthesize novel views of unseen scenes. Our method is capable of predicting the color of a target ray in a novel scene directly, just from a collection of patches sampled from the scene. We show that our approach outperforms the state-of-the-art on novel view synthesis of unseen scenes even when being trained with considerably less data than prior work.
  arXiv  Detail & Related papers  (2022-07-21T17:57:04Z)
• R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis [76.07010495581535]
  Rendering a single pixel requires querying the Neural Radiance Field network hundreds of times. NeLF presents a more straightforward representation over NeRF in novel view. We show the key to successfully learning a deep NeLF network is to have sufficient data.
  arXiv  Detail & Related papers  (2022-03-31T17:57:05Z)
• NeRF-SR: High-Quality Neural Radiance Fields using Super-Sampling [82.99453001445478]
  We present NeRF-SR, a solution for high-resolution (HR) novel view synthesis with mostly low-resolution (LR) inputs. Our method is built upon Neural Radiance Fields (NeRF) that predicts per-point density and color with a multi-layer perceptron.
  arXiv  Detail & Related papers  (2021-12-03T07:33:47Z)

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.