Not-So-Optimal Transport Flows for 3D Point Cloud Generation

• URL: http://arxiv.org/abs/2502.12456v1
• Date: Tue, 18 Feb 2025 02:37:34 GMT
• Title: Not-So-Optimal Transport Flows for 3D Point Cloud Generation
• Authors: Ka-Hei Hui, Chao Liu, Xiaohui Zeng, Chi-Wing Fu, Arash Vahdat,
• Abstract summary: Learning generative models of 3D point clouds is one of the fundamental problems in 3D generative learning. In this paper, we analyze the recently proposed equivariant OT flows that learn permutation invariant generative models for point-based molecular data. We show that our proposed model outperforms prior diffusion- and flow-based approaches on a wide range of unconditional generation and shape completion.
• Score: 58.164908756416615
• License:
• Abstract: Learning generative models of 3D point clouds is one of the fundamental problems in 3D generative learning. One of the key properties of point clouds is their permutation invariance, i.e., changing the order of points in a point cloud does not change the shape they represent. In this paper, we analyze the recently proposed equivariant OT flows that learn permutation invariant generative models for point-based molecular data and we show that these models scale poorly on large point clouds. Also, we observe learning (equivariant) OT flows is generally challenging since straightening flow trajectories makes the learned flow model complex at the beginning of the trajectory. To remedy these, we propose not-so-optimal transport flow models that obtain an approximate OT by an offline OT precomputation, enabling an efficient construction of OT pairs for training. During training, we can additionally construct a hybrid coupling by combining our approximate OT and independent coupling to make the target flow models easier to learn. In an extensive empirical study, we show that our proposed model outperforms prior diffusion- and flow-based approaches on a wide range of unconditional generation and shape completion on the ShapeNet benchmark.

Related papers

• Equi-GSPR: Equivariant SE(3) Graph Network Model for Sparse Point Cloud Registration [2.814748676983944]
  We propose a graph neural network model embedded with a local Spherical Euclidean 3D equivariance property through SE(3) message passing based propagation. Our model is composed mainly of a descriptor module, equivariant graph layers, match similarity, and the final regression layers. Experiments conducted on the 3DMatch and KITTI datasets exhibit the compelling and robust performance of our model compared to state-of-the-art approaches.
  arXiv  Detail & Related papers  (2024-10-08T06:48:01Z)
• DiffFacto: Controllable Part-Based 3D Point Cloud Generation with Cross Diffusion [68.39543754708124]
  We introduce DiffFacto, a novel probabilistic generative model that learns the distribution of shapes with part-level control. Experiments show that our method is able to generate novel shapes with multiple axes of control. It achieves state-of-the-art part-level generation quality and generates plausible and coherent shapes.
  arXiv  Detail & Related papers  (2023-05-03T06:38:35Z)
• StarNet: Style-Aware 3D Point Cloud Generation [82.30389817015877]
  StarNet is able to reconstruct and generate high-fidelity and even 3D point clouds using a mapping network. Our framework achieves comparable state-of-the-art performance on various metrics in the point cloud reconstruction and generation tasks.
  arXiv  Detail & Related papers  (2023-03-28T08:21:44Z)
• SCOOP: Self-Supervised Correspondence and Optimization-Based Scene Flow [25.577386156273256]
  Scene flow estimation is a long-standing problem in computer vision, where the goal is to find the 3D motion of a scene from its consecutive observations. We introduce SCOOP, a new method for scene flow estimation that can be learned on a small amount of data without employing ground-truth flow supervision.
  arXiv  Detail & Related papers  (2022-11-25T10:52:02Z)
• Learned Vertex Descent: A New Direction for 3D Human Model Fitting [64.04726230507258]
  We propose a novel optimization-based paradigm for 3D human model fitting on images and scans. Our approach is able to capture the underlying body of clothed people with very different body shapes, achieving a significant improvement compared to state-of-the-art. LVD is also applicable to 3D model fitting of humans and hands, for which we show a significant improvement to the SOTA with a much simpler and faster method.
  arXiv  Detail & Related papers  (2022-05-12T17:55:51Z)
• PU-Flow: a Point Cloud Upsampling Networkwith Normalizing Flows [58.96306192736593]
  We present PU-Flow, which incorporates normalizing flows and feature techniques to produce dense points uniformly distributed on the underlying surface. Specifically, we formulate the upsampling process as point in a latent space, where the weights are adaptively learned from local geometric context. We show that our method outperforms state-of-the-art deep learning-based approaches in terms of reconstruction quality, proximity-to-surface accuracy, and computation efficiency.
  arXiv  Detail & Related papers  (2021-07-13T07:45:48Z)
• Go with the Flows: Mixtures of Normalizing Flows for Point Cloud Generation and Reconstruction [98.38585659305325]
  normalizing flows (NFs) have demonstrated state-of-the-art performance on modeling 3D point clouds. This work enhances their representational power by applying mixtures of NFs to point clouds.
  arXiv  Detail & Related papers  (2021-06-06T14:25:45Z)
• Continual Learning of Generative Models with Limited Data: From Wasserstein-1 Barycenter to Adaptive Coalescence [22.82926450287203]
  Learning generative models is challenging for a network edge node with limited data and computing power. This study aims to develop a framework which systematically optimize continual learning of generative models.
  arXiv  Detail & Related papers  (2021-01-22T17:15:39Z)
• Discrete Point Flow Networks for Efficient Point Cloud Generation [36.03093265136374]
  Generative models have proven effective at modeling 3D shapes and their statistical variations. We introduce a latent variable model that builds on normalizing flows to generate 3D point clouds of an arbitrary size. For single-view shape reconstruction we also obtain results on par with state-of-the-art voxel, point cloud, and mesh-based methods.
  arXiv  Detail & Related papers  (2020-07-20T14:48:00Z)

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.