F-HOI: Toward Fine-grained Semantic-Aligned 3D Human-Object Interactions
This addresses the need for detailed understanding of intermediate states and transitions in HOI for computer vision and robotics applications, representing an incremental advancement with a novel dataset and unified model.
The paper tackles the problem of coarse alignment in 3D human-object interaction (HOI) datasets and models by introducing fine-grained semantic alignment using state-level descriptions, resulting in a new dataset (Semantic-HOI) with over 20K paired HOI states and a unified model (F-HOI) that effectively handles diverse HOI tasks.
Existing 3D human object interaction (HOI) datasets and models simply align global descriptions with the long HOI sequence, while lacking a detailed understanding of intermediate states and the transitions between states. In this paper, we argue that fine-grained semantic alignment, which utilizes state-level descriptions, offers a promising paradigm for learning semantically rich HOI representations. To achieve this, we introduce Semantic-HOI, a new dataset comprising over 20K paired HOI states with fine-grained descriptions for each HOI state and the body movements that happen between two consecutive states. Leveraging the proposed dataset, we design three state-level HOI tasks to accomplish fine-grained semantic alignment within the HOI sequence. Additionally, we propose a unified model called F-HOI, designed to leverage multimodal instructions and empower the Multi-modal Large Language Model to efficiently handle diverse HOI tasks. F-HOI offers multiple advantages: (1) It employs a unified task formulation that supports the use of versatile multimodal inputs. (2) It maintains consistency in HOI across 2D, 3D, and linguistic spaces. (3) It utilizes fine-grained textual supervision for direct optimization, avoiding intricate modeling of HOI states. Extensive experiments reveal that F-HOI effectively aligns HOI states with fine-grained semantic descriptions, adeptly tackling understanding, reasoning, generation, and reconstruction tasks.