PAVLM: Advancing Point Cloud based Affordance Understanding Via Vision-Language Model

TL;DR

PAVLM is a novel framework that enhances 3D affordance understanding in point clouds by integrating large language models with geometric modules, improving generalization in open-world robotic tasks.

Contribution

The paper introduces PAVLM, combining pre-trained language models with geometric-guided propagation to advance 3D affordance understanding from point clouds.

Findings

Outperforms baseline methods on 3D-AffordanceNet benchmark.

Excels in generalizing to novel open-world affordance tasks.

Enhances semantic understanding of physical properties in 3D objects.

Abstract

Affordance understanding, the task of identifying actionable regions on 3D objects, plays a vital role in allowing robotic systems to engage with and operate within the physical world. Although Visual Language Models (VLMs) have excelled in high-level reasoning and long-horizon planning for robotic manipulation, they still fall short in grasping the nuanced physical properties required for effective human-robot interaction. In this paper, we introduce PAVLM (Point cloud Affordance Vision-Language Model), an innovative framework that utilizes the extensive multimodal knowledge embedded in pre-trained language models to enhance 3D affordance understanding of point cloud. PAVLM integrates a geometric-guided propagation module with hidden embeddings from large language models (LLMs) to enrich visual semantics. On the language side, we prompt Llama-3.1 models to generate refined context-aware text, augmenting the instructional input with deeper semantic cues. Experimental results on the 3D-AffordanceNet benchmark demonstrate that PAVLM outperforms baseline methods for both full and partial point clouds, particularly excelling in its generalization to novel open-world affordance tasks of 3D objects. For more information, visit our project site: pavlm-source.github.io.