Self-supervised Light Field View Synthesis Using Cycle Consistency

• URL: http://arxiv.org/abs/2008.05084v1
• Date: Wed, 12 Aug 2020 03:20:19 GMT
• Title: Self-supervised Light Field View Synthesis Using Cycle Consistency
• Authors: Yang Chen, Martin Alain, Aljosa Smolic
• Abstract summary: We propose a self-supervised light field view synthesis framework with cycle consistency. A cycle consistency constraint is used to build mapping enforcing the generated views to be consistent with the input views. Results show it outperforms state-of-the-art light field view synthesis methods, especially when generating multiple intermediate views.
• Score: 22.116100469958436
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: High angular resolution is advantageous for practical applications of light fields. In order to enhance the angular resolution of light fields, view synthesis methods can be utilized to generate dense intermediate views from sparse light field input. Most successful view synthesis methods are learning-based approaches which require a large amount of training data paired with ground truth. However, collecting such large datasets for light fields is challenging compared to natural images or videos. To tackle this problem, we propose a self-supervised light field view synthesis framework with cycle consistency. The proposed method aims to transfer prior knowledge learned from high quality natural video datasets to the light field view synthesis task, which reduces the need for labeled light field data. A cycle consistency constraint is used to build bidirectional mapping enforcing the generated views to be consistent with the input views. Derived from this key concept, two loss functions, cycle loss and reconstruction loss, are used to fine-tune the pre-trained model of a state-of-the-art video interpolation method. The proposed method is evaluated on various datasets to validate its robustness, and results show it not only achieves competitive performance compared to supervised fine-tuning, but also outperforms state-of-the-art light field view synthesis methods, especially when generating multiple intermediate views. Besides, our generic light field view synthesis framework can be adopted to any pre-trained model for advanced video interpolation.

Related papers

• UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting [85.27994475113056]
  We introduce a general-purpose approach that jointly estimates albedo and synthesizes relit outputs in a single pass.<n>Our model demonstrates strong generalization across diverse domains and surpasses previous methods in both visual fidelity and temporal consistency.
  arXiv  Detail & Related papers  (2025-06-18T17:56:45Z)
• Rendering Anywhere You See: Renderability Field-guided Gaussian Splatting [4.89907242398523]
  We propose renderability field-guided gaussian splatting (RF-GS) for scene view synthesis. RF-GS quantifies input inhomogeneity through a renderability field, guiding pseudo-view sampling to enhanced visual consistency. Our experiments on simulated and real-world data show that our method outperforms existing approaches in rendering stability.
  arXiv  Detail & Related papers  (2025-04-27T14:41:01Z)
• Synthesizing Consistent Novel Views via 3D Epipolar Attention without Re-Training [102.82553402539139]
  Large diffusion models demonstrate remarkable zero-shot capabilities in novel view synthesis from a single image. These models often face challenges in maintaining consistency across novel and reference views. We propose to use epipolar geometry to locate and retrieve overlapping information from the input view. This information is then incorporated into the generation of target views, eliminating the need for training or fine-tuning.
  arXiv  Detail & Related papers  (2025-02-25T14:04:22Z)
• Relighting from a Single Image: Datasets and Deep Intrinsic-based Architecture [0.7499722271664147]
  Single image scene relighting aims to generate a realistic new version of an input image so that it appears to be illuminated by a new target light condition. We propose two new datasets: a synthetic dataset with the ground truth of intrinsic components and a real dataset collected under laboratory conditions. Our method outperforms the state-of-the-art methods in performance, as tested on both existing datasets and our newly developed datasets.
  arXiv  Detail & Related papers  (2024-09-27T14:15:02Z)
• ViewFusion: Learning Composable Diffusion Models for Novel View Synthesis [47.57948804514928]
  This work introduces ViewFusion, a state-of-the-art end-to-end generative approach to novel view synthesis. ViewFusion consists in simultaneously applying a diffusion denoising step to any number of input views of a scene.
  arXiv  Detail & Related papers  (2024-02-05T11:22:14Z)
• Layered Rendering Diffusion Model for Zero-Shot Guided Image Synthesis [60.260724486834164]
  This paper introduces innovative solutions to enhance spatial controllability in diffusion models reliant on text queries. We present two key innovations: Vision Guidance and the Layered Rendering Diffusion framework. We apply our method to three practical applications: bounding box-to-image, semantic mask-to-image and image editing.
  arXiv  Detail & Related papers  (2023-11-30T10:36:19Z)
• Reconstructing Continuous Light Field From Single Coded Image [7.937367109582907]
  We propose a method for reconstructing a continuous light field of a target scene from a single observed image. Joint aperture-exposure coding implemented in a camera enables effective embedding of 3-D scene information into an observed image. NeRF-based neural rendering enables high quality view synthesis of a 3-D scene from continuous viewpoints.
  arXiv  Detail & Related papers  (2023-11-16T07:59:01Z)
• ContraNeRF: Generalizable Neural Radiance Fields for Synthetic-to-real Novel View Synthesis via Contrastive Learning [102.46382882098847]
  We first investigate the effects of synthetic data in synthetic-to-real novel view synthesis. We propose to introduce geometry-aware contrastive learning to learn multi-view consistent features with geometric constraints. Our method can render images with higher quality and better fine-grained details, outperforming existing generalizable novel view synthesis methods in terms of PSNR, SSIM, and LPIPS.
  arXiv  Detail & Related papers  (2023-03-20T12:06:14Z)
• HORIZON: High-Resolution Semantically Controlled Panorama Synthesis [105.55531244750019]
  Panorama synthesis endeavors to craft captivating 360-degree visual landscapes, immersing users in the heart of virtual worlds. Recent breakthroughs in visual synthesis have unlocked the potential for semantic control in 2D flat images, but a direct application of these methods to panorama synthesis yields distorted content. We unveil an innovative framework for generating high-resolution panoramas, adeptly addressing the issues of spherical distortion and edge discontinuity through sophisticated spherical modeling.
  arXiv  Detail & Related papers  (2022-10-10T09:43:26Z)
• Progressively-connected Light Field Network for Efficient View Synthesis [69.29043048775802]
  We present a Progressively-connected Light Field network (ProLiF) for the novel view synthesis of complex forward-facing scenes. ProLiF encodes a 4D light field, which allows rendering a large batch of rays in one training step for image- or patch-level losses.
  arXiv  Detail & Related papers  (2022-07-10T13:47:20Z)
• Content-aware Warping for View Synthesis [110.54435867693203]
  We propose content-aware warping, which adaptively learns the weights for pixels of a relatively large neighborhood from their contextual information via a lightweight neural network. Based on this learnable warping module, we propose a new end-to-end learning-based framework for novel view synthesis from two source views. Experimental results on structured light field datasets with wide baselines and unstructured multi-view datasets show that the proposed method significantly outperforms state-of-the-art methods both quantitatively and visually.
  arXiv  Detail & Related papers  (2022-01-22T11:35:05Z)
• Light Field Neural Rendering [47.7586443731997]
  Methods based on geometric reconstruction need only sparse views, but cannot accurately model non-Lambertian effects. We introduce a model that combines the strengths and mitigates the limitations of these two directions. Our model outperforms the state-of-the-art on multiple forward-facing and 360deg datasets.
  arXiv  Detail & Related papers  (2021-12-17T18:58:05Z)
• Learning optical flow from still images [53.295332513139925]
  We introduce a framework to generate accurate ground-truth optical flow annotations quickly and in large amounts from any readily available single real picture. We virtually move the camera in the reconstructed environment with known motion vectors and rotation angles. When trained with our data, state-of-the-art optical flow networks achieve superior generalization to unseen real data.
  arXiv  Detail & Related papers  (2021-04-08T17:59:58Z)

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.