Pixel-Aligned Multi-View Generation with Depth Guided Decoder

• URL: http://arxiv.org/abs/2408.14016v1
• Date: Mon, 26 Aug 2024 04:56:41 GMT
• Title: Pixel-Aligned Multi-View Generation with Depth Guided Decoder
• Authors: Zhenggang Tang, Peiye Zhuang, Chaoyang Wang, Aliaksandr Siarohin, Yash Kant, Alexander Schwing, Sergey Tulyakov, Hsin-Ying Lee,
• Abstract summary: We propose a novel method for pixel-level image-to-multi-view generation. Unlike prior work, we incorporate attention layers across multi-view images in the VAE decoder of a latent video diffusion model. Our model enables better pixel alignment across multi-view images.
• Score: 86.1813201212539
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The task of image-to-multi-view generation refers to generating novel views of an instance from a single image. Recent methods achieve this by extending text-to-image latent diffusion models to multi-view version, which contains an VAE image encoder and a U-Net diffusion model. Specifically, these generation methods usually fix VAE and finetune the U-Net only. However, the significant downscaling of the latent vectors computed from the input images and independent decoding leads to notable pixel-level misalignment across multiple views. To address this, we propose a novel method for pixel-level image-to-multi-view generation. Unlike prior work, we incorporate attention layers across multi-view images in the VAE decoder of a latent video diffusion model. Specifically, we introduce a depth-truncated epipolar attention, enabling the model to focus on spatially adjacent regions while remaining memory efficient. Applying depth-truncated attn is challenging during inference as the ground-truth depth is usually difficult to obtain and pre-trained depth estimation models is hard to provide accurate depth. Thus, to enhance the generalization to inaccurate depth when ground truth depth is missing, we perturb depth inputs during training. During inference, we employ a rapid multi-view to 3D reconstruction approach, NeuS, to obtain coarse depth for the depth-truncated epipolar attention. Our model enables better pixel alignment across multi-view images. Moreover, we demonstrate the efficacy of our approach in improving downstream multi-view to 3D reconstruction tasks.

Related papers

• Era3D: High-Resolution Multiview Diffusion using Efficient Row-wise Attention [87.02613021058484]
  We introduce Era3D, a novel multiview diffusion method that generates high-resolution multiview images from a single-view image. Era3D generates high-quality multiview images with up to a 512*512 resolution while reducing complexity by 12x times.
  arXiv  Detail & Related papers  (2024-05-19T17:13:16Z)
• MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation [54.27399121779011]
  We present MVD-Fusion: a method for single-view 3D inference via generative modeling of multi-view-consistent RGB-D images. We show that our approach can yield more accurate synthesis compared to recent state-of-the-art, including distillation-based 3D inference and prior multi-view generation methods.
  arXiv  Detail & Related papers  (2024-04-04T17:59:57Z)
• Envision3D: One Image to 3D with Anchor Views Interpolation [18.31796952040799]
  We present Envision3D, a novel method for efficiently generating high-quality 3D content from a single image. It is capable of generating high-quality 3D content in terms of texture and geometry, surpassing previous image-to-3D baseline methods.
  arXiv  Detail & Related papers  (2024-03-13T18:46:33Z)
• Hyper-VolTran: Fast and Generalizable One-Shot Image to 3D Object Structure via HyperNetworks [53.67497327319569]
  We introduce a novel neural rendering technique to solve image-to-3D from a single view. Our approach employs the signed distance function as the surface representation and incorporates generalizable priors through geometry-encoding volumes and HyperNetworks. Our experiments show the advantages of our proposed approach with consistent results and rapid generation.
  arXiv  Detail & Related papers  (2023-12-24T08:42:37Z)
• Neural Radiance Fields Approach to Deep Multi-View Photometric Stereo [103.08512487830669]
  We present a modern solution to the multi-view photometric stereo problem (MVPS) We procure the surface orientation using a photometric stereo (PS) image formation model and blend it with a multi-view neural radiance field representation to recover the object's surface geometry. Our method performs neural rendering of multi-view images while utilizing surface normals estimated by a deep photometric stereo network.
  arXiv  Detail & Related papers  (2021-10-11T20:20:03Z)
• DeepMultiCap: Performance Capture of Multiple Characters Using Sparse Multiview Cameras [63.186486240525554]
  DeepMultiCap is a novel method for multi-person performance capture using sparse multi-view cameras. Our method can capture time varying surface details without the need of using pre-scanned template models.
  arXiv  Detail & Related papers  (2021-05-01T14:32:13Z)
• Multimodal Deep Unfolding for Guided Image Super-Resolution [23.48305854574444]
  Deep learning methods rely on training data to learn an end-to-end mapping from a low-resolution input to a high-resolution output. We propose a multimodal deep learning design that incorporates sparse priors and allows the effective integration of information from another image modality into the network architecture. Our solution relies on a novel deep unfolding operator, performing steps similar to an iterative algorithm for convolutional sparse coding with side information.
  arXiv  Detail & Related papers  (2020-01-21T14:41:53Z)

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.