Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following

TL;DR

This paper introduces Point-Bind, a multi-modal 3D model aligning point clouds with various modalities, and Point-LLM, a 3D large language model capable of understanding and following multi-modal instructions, advancing 3D understanding and generation.

Contribution

The work presents the first multi-modal 3D point cloud model and a 3D large language model that leverages multi-modal alignment without requiring 3D instruction data.

Findings

Point-Bind effectively aligns 3D data with multiple modalities.

Point-LLM demonstrates superior 3D question-answering capabilities.

The models enable new applications like any-to-3D generation and 3D embedding arithmetic.

Abstract

We introduce Point-Bind, a 3D multi-modality model aligning point clouds with 2D image, language, audio, and video. Guided by ImageBind, we construct a joint embedding space between 3D and multi-modalities, enabling many promising applications, e.g., any-to-3D generation, 3D embedding arithmetic, and 3D open-world understanding. On top of this, we further present Point-LLM, the first 3D large language model (LLM) following 3D multi-modal instructions. By parameter-efficient fine-tuning techniques, Point-LLM injects the semantics of Point-Bind into pre-trained LLMs, e.g., LLaMA, which requires no 3D instruction data, but exhibits superior 3D and multi-modal question-answering capacity. We hope our work may cast a light on the community for extending 3D point clouds to multi-modality applications. Code is available at https://github.com/ZiyuGuo99/Point-Bind_Point-LLM.