Search papers, labs, and topics across Lattice.
The paper introduces HY3D-Bench, a new open-source ecosystem for 3D generation, comprising a curated library of 250k high-fidelity 3D objects and 125k synthetic assets. The dataset addresses data processing bottlenecks in 3D content creation by providing training-ready artifacts with watertight meshes, multi-view renderings, and structured part-level decomposition. Empirical validation through training Hunyuan3D-2.1-Small demonstrates the dataset's utility in advancing 3D perception, robotics, and digital content creation.
Unlock controllable 3D asset generation with HY3D-Bench, a meticulously curated and augmented dataset of 375k objects featuring part-level decomposition.
While recent advances in neural representations and generative models have revolutionized 3D content creation, the field remains constrained by significant data processing bottlenecks. To address this, we introduce HY3D-Bench, an open-source ecosystem designed to establish a unified, high-quality foundation for 3D generation. Our contributions are threefold: (1) We curate a library of 250k high-fidelity 3D objects distilled from large-scale repositories, employing a rigorous pipeline to deliver training-ready artifacts, including watertight meshes and multi-view renderings; (2) We introduce structured part-level decomposition, providing the granularity essential for fine-grained perception and controllable editing; and (3) We bridge real-world distribution gaps via a scalable AIGC synthesis pipeline, contributing 125k synthetic assets to enhance diversity in long-tail categories. Validated empirically through the training of Hunyuan3D-2.1-Small, HY3D-Bench democratizes access to robust data resources, aiming to catalyze innovation across 3D perception, robotics, and digital content creation.
