Search papers, labs, and topics across Lattice.
The paper introduces TeHOR, a framework for joint 3D human and object reconstruction from a single image that addresses limitations of existing methods by incorporating text descriptions of human-object interactions and appearance cues. TeHOR enforces semantic alignment between the 3D reconstruction and textual cues, enabling reasoning about both contact and non-contact interactions. By integrating appearance cues, the framework captures holistic contextual information, leading to visually plausible reconstructions.
Reconstructing 3D scenes just got a whole lot smarter: TeHOR uses text descriptions to understand human-object interactions beyond physical contact, leading to more accurate and contextually aware reconstructions.
Joint reconstruction of 3D human and object from a single image is an active research area, with pivotal applications in robotics and digital content creation. Despite recent advances, existing approaches suffer from two fundamental limitations. First, their reconstructions rely heavily on physical contact information, which inherently cannot capture non-contact human-object interactions, such as gazing at or pointing toward an object. Second, the reconstruction process is primarily driven by local geometric proximity, neglecting the human and object appearances that provide global context crucial for understanding holistic interactions. To address these issues, we introduce TeHOR, a framework built upon two core designs. First, beyond contact information, our framework leverages text descriptions of human-object interactions to enforce semantic alignment between the 3D reconstruction and its textual cues, enabling reasoning over a wider spectrum of interactions, including non-contact cases. Second, we incorporate appearance cues of the 3D human and object into the alignment process to capture holistic contextual information, thereby ensuring visually plausible reconstructions. As a result, our framework produces accurate and semantically coherent reconstructions, achieving state-of-the-art performance.
