EI-Part: Explode for Completion and Implode for Refinement

Abstract

Part-level 3D generation is crucial for various downstream applications, including gaming, film production, and industrial design. However, decomposing a 3D shape into geometrically plausible and meaningful components remains a significant challenge. Previous part-based generation methods often struggle to produce well-constructed parts, exhibiting poor structural coherence, geometric implausibility, inaccuracy, or inefficiency. To address these challenges, we introduce EI-Part, a novel framework specifically designed to generate high-quality 3D shapes with components, characterized by strong structural coherence, geometric plausibility, geometric fidelity, and generation efficiency. We propose utilizing distinct representations at different stages: an Explode state for part completion and an Implode state for geometry refinement. This strategy fully leverages spatial resolution, enabling flexible part completion and fine geometric detail generation. To maintain structural coherence between parts, a self-attention mechanism is incorporated in both exploded and imploded states, facilitating effective information perception and feature fusion among components during generation. Extensive experiments on multiple benchmarks demonstrate that EI-Part efficiently produces semantically meaningful and structurally coherent parts with fine-grained geometric details, achieving state-of-the-art performance in part-level 3D generation. Project page: https://cvhadessun.github.io/EI-Part/

Figures (7)

• Figure 1: We present EI-Part, a novel framework designed to efficiently produce structurally coherent and semantically meaningful parts with fine-grained geometric details. Our work proposes an Explode-Implode strategy to fully utilize spatial resolution at different stages for meaningful part completion and geometric detail generation.
• Figure 2: The pipeline of our proposed EI-Part. Given an input 3D shape $O$, we first obtain its normal map $\{n_i\}_{i=1}^6$ and canonical coordinate maps (CCMs) $\{c_i\}_{i=1}^6$ from six views. We then perform frontal segmentation using SAM and employ MVSegNet to achieve multi-view consistent segmentations. These segmentations are lifted to 3D to create an initial part segmentation, which is subsequently inpainted by InSegNet to produce accurate 3D segmented shapes $\{p_{s}^k\}_{k=1}^K$. The segmented result is exploded into discrete 3D voxels, enabling us to perform conditional diffusion completion to generate geometrically plausible and complete part structures $e_c$. Next, the exploded complete parts are imploded back to a compact state for conditional diffusion refinement, capturing fine-level details $g_d$. Thanks to the utilization of the exploded-imploded strategy, we fully utilize spatial resolution at different stages, enhancing the accuracy of part completion and surface details. As a result, the 3D shapes $g_d$ generated by our method exhibit individual, structurally consistent, and geometrically plausible part shapes with fine-grained geometric details.
• Figure 3: Qualitative comparison between ours and the state of the arts. For each row, the input model (a) is followed by the parts generated by (b) BANG zhang2025bang, (c) HoloPart yang2025holopart, (d) X-Part yan2025x, (e) OmniPart yang2025omnipart and (f) our method. Our method demonstrates proficiency in generating high-quality 3D shapes with well-defined parts, achieving notable performance in structural coherence, geometric plausibility, fidelity, and efficiency. Please refer to our supplementary materials for a qualitative comparison with additional methods (e.g., PartPacker and PartCrafter).
• Figure 4: Qualitative comparison between our method and the state of the art, where all methods share the segmentation results of ours. Each row presents the input model (a), followed by (b) our segmentation, and the parts generated by (c) HoloPart yang2025holopart, (d) X-Part yan2025x, (e) OmniPart yang2025omnipart and (f) our method.
• Figure 5: Qualitative comparison between our method and the alternative strategies. The input is the same segmentation.