Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

training-free text-guided point cloud generation

Training-free text-guided point cloud generation refers to the process of creating three-dimensional point clouds that match natural language descriptions without requiring neural network training or fine-tuning on paired text-and-3D datasets. Instead of learning direct text-to-shape mappings through supervised training, this approach leverages pre-trained unconditional 3D generative models alongside pre-trained vision-language or multimodal representations. During the sampling or inference phase, semantic signals from a textual prompt are applied at test time to guide and adjust the generative trajectory, producing shapes that align with the text description while bypassing the computational cost and data scarcity associated with training specialized text-to-3D models.

1 item

Fast Point Cloud Generation with Straight Flows

Fast Point Cloud Generation with Straight Flows

Lemeng Wu, Dilin Wang, Chengyue Gong, Xingchao Liu, Yunyang Xiong, Rakesh Ranjan, Raghuraman Krishnamoorthi, Vikas Chandra, Qiang Liu

OrganizationsMetaUniversity of Texas at Austin

Why you should read this

Proposes Point Straight Flow, a novel framework that straightens generative transport trajectories and distills them into a single step, enabling high-quality 3D point cloud generation over 700 times faster than standard diffusion models.

Diffusion models have emerged as a powerful tool for point cloud generation. A key component that drives the impressive performance for generating high-quality samples from noise is iteratively denoise for thousands of steps. While beneficial, the complexity of learning steps has limited its applications to many 3D real-world. To address this limitation, we propose Point Straight Flow (PSF), a model that exhibits impressive performance using one step. Our idea is based on the reformulation of the standard diffusion model, which optimizes the curvy learning trajectory into a straight path. Further, we develop a distillation strategy to shorten the straight path into one step without a performance loss, enabling applications to 3D real-world with latency constraints. We perform evaluations on multiple 3D tasks and find that our PSF performs comparably to the standard diffusion model, outperforming other efficient 3D point cloud generation methods. On real-world applications such as point cloud completion and training-free text-guided generation in a low-latency setup, PSF performs favorably.

Added

2026-09-26