DiT-3D: Exploring Plain Diffusion Transformers for 3D Shape Generation

• URL: http://arxiv.org/abs/2307.01831v1
• Date: Tue, 4 Jul 2023 17:15:46 GMT
• Title: DiT-3D: Exploring Plain Diffusion Transformers for 3D Shape Generation
• Authors: Shentong Mo, Enze Xie, Ruihang Chu, Lewei Yao, Lanqing Hong, Matthias Nie{\ss}ner, Zhenguo Li
• Abstract summary: We propose a novel Diffusion Transformer for 3D shape generation, namely DiT-3D. Compared to existing U-Net approaches, our DiT-3D is more scalable in model size and produces much higher quality generations. Experimental results on the ShapeNet dataset demonstrate that the proposed DiT-3D achieves state-of-the-art performance.
• Score: 49.22974835756199
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent Diffusion Transformers (e.g., DiT) have demonstrated their powerful effectiveness in generating high-quality 2D images. However, it is still being determined whether the Transformer architecture performs equally well in 3D shape generation, as previous 3D diffusion methods mostly adopted the U-Net architecture. To bridge this gap, we propose a novel Diffusion Transformer for 3D shape generation, namely DiT-3D, which can directly operate the denoising process on voxelized point clouds using plain Transformers. Compared to existing U-Net approaches, our DiT-3D is more scalable in model size and produces much higher quality generations. Specifically, the DiT-3D adopts the design philosophy of DiT but modifies it by incorporating 3D positional and patch embeddings to adaptively aggregate input from voxelized point clouds. To reduce the computational cost of self-attention in 3D shape generation, we incorporate 3D window attention into Transformer blocks, as the increased 3D token length resulting from the additional dimension of voxels can lead to high computation. Finally, linear and devoxelization layers are used to predict the denoised point clouds. In addition, our transformer architecture supports efficient fine-tuning from 2D to 3D, where the pre-trained DiT-2D checkpoint on ImageNet can significantly improve DiT-3D on ShapeNet. Experimental results on the ShapeNet dataset demonstrate that the proposed DiT-3D achieves state-of-the-art performance in high-fidelity and diverse 3D point cloud generation. In particular, our DiT-3D decreases the 1-Nearest Neighbor Accuracy of the state-of-the-art method by 4.59 and increases the Coverage metric by 3.51 when evaluated on Chamfer Distance.

Related papers

• Deep Geometric Moments Promote Shape Consistency in Text-to-3D Generation [27.43973967994717]
  MT3D is a text-to-3D generative model that leverages a high-fidelity 3D object to overcome viewpoint bias. We employ depth maps derived from a high-quality 3D model as control signals to guarantee that the generated 2D images preserve the fundamental shape and structure. By incorporating geometric details from a 3D asset, MT3D enables the creation of diverse and geometrically consistent objects.
  arXiv  Detail & Related papers  (2024-08-12T06:25:44Z)
• DIRECT-3D: Learning Direct Text-to-3D Generation on Massive Noisy 3D Data [50.164670363633704]
  We present DIRECT-3D, a diffusion-based 3D generative model for creating high-quality 3D assets from text prompts. Our model is directly trained on extensive noisy and unaligned in-the-wild' 3D assets. We achieve state-of-the-art performance in both single-class generation and text-to-3D generation.
  arXiv  Detail & Related papers  (2024-06-06T17:58:15Z)
• Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformer [26.375689838055774]
  Direct3D is a native 3D generative model scalable to in-the-wild input images. Our approach comprises two primary components: a Direct 3D Variational Auto-Encoder (D3D-VAE) and a Direct 3D Diffusion Transformer (D3D-DiT)
  arXiv  Detail & Related papers  (2024-05-23T17:49:37Z)
• Any2Point: Empowering Any-modality Large Models for Efficient 3D Understanding [83.63231467746598]
  We introduce Any2Point, a parameter-efficient method to empower any-modality large models (vision, language, audio) for 3D understanding. We propose a 3D-to-any (1D or 2D) virtual projection strategy that correlates the input 3D points to the original 1D or 2D positions within the source modality.
  arXiv  Detail & Related papers  (2024-04-11T17:59:45Z)
• LN3Diff: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation [73.36690511083894]
  This paper introduces a novel framework called LN3Diff to address a unified 3D diffusion pipeline. Our approach harnesses a 3D-aware architecture and variational autoencoder to encode the input image into a structured, compact, and 3D latent space. It achieves state-of-the-art performance on ShapeNet for 3D generation and demonstrates superior performance in monocular 3D reconstruction and conditional 3D generation.
  arXiv  Detail & Related papers  (2024-03-18T17:54:34Z)
• IM-3D: Iterative Multiview Diffusion and Reconstruction for High-Quality 3D Generation [96.32684334038278]
  In this paper, we explore the design space of text-to-3D models. We significantly improve multi-view generation by considering video instead of image generators. Our new method, IM-3D, reduces the number of evaluations of the 2D generator network 10-100x.
  arXiv  Detail & Related papers  (2024-02-13T18:59:51Z)
• Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors [16.93758384693786]
  Bidirectional Diffusion(BiDiff) is a unified framework that incorporates both a 3D and a 2D diffusion process. Our model achieves high-quality, diverse, and scalable 3D generation.
  arXiv  Detail & Related papers  (2023-12-07T10:00:04Z)
• Points-to-3D: Bridging the Gap between Sparse Points and Shape-Controllable Text-to-3D Generation [16.232803881159022]
  We propose a flexible framework of Points-to-3D to bridge the gap between sparse yet freely available 3D points and realistic shape-controllable 3D generation. The core idea of Points-to-3D is to introduce controllable sparse 3D points to guide the text-to-3D generation.
  arXiv  Detail & Related papers  (2023-07-26T02:16:55Z)
• NeRF-GAN Distillation for Efficient 3D-Aware Generation with Convolutions [97.27105725738016]
  integration of Neural Radiance Fields (NeRFs) and generative models, such as Generative Adversarial Networks (GANs) has transformed 3D-aware generation from single-view images. We propose a simple and effective method, based on re-using the well-disentangled latent space of a pre-trained NeRF-GAN in a pose-conditioned convolutional network to directly generate 3D-consistent images corresponding to the underlying 3D representations.
  arXiv  Detail & Related papers  (2023-03-22T18:59:48Z)

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.