Newest papers

Gate3DGen: A Pre-processing Model that Selectively Purifies Real-World Images for 3D Generation

Xiaolin Li, Zihao Huang, Fanpeng Meng, Chen Yang, Guanjun Wu, Kai Zhang, Jiemin Fang, Qi Tian · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31649 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Generasi 3D single-image correction telah mencapai loyalitas tinggi pada referensi curated, namun degradasi tajam pada gambar dunia nyata dengan pencahayaan yang tidak merata, kabur, dan tertutup. Pembersihan input semacam itu membutuhkan keputusan kapan untuk campur tangan dan mempertahankan petunjuk visual yang diperlukan untuk generasi 3D, sementara meninggalkan input bersih tidak berubah. Kami memperkenalkan Gate3DGen, model pra-proses yang selektif yang mengarahkan input gambar ke generator 3D yang ada tanpa melatih mereka kembali. Router sadar degradasi melewati input yang diklasifikasikan sebagai bersih dan mengekstrak fitur untuk membimbing pembersihan input yang tersisa. Disesuaikan dengan fitur-fitur ini, pembersih satu langkah 3D memurnikan input yang degradasi dengan kerugian spasial yang mendorong konservasi identitas objek, kontur

Machine translation (Meta M2M100); the English original follows.

Abstract

Single-image 3D generation has achieved high fidelity on curated benchmarks, yet degrades sharply on real-world images with uneven illumination, blur, and occlusions. Purifying such inputs requires deciding when to intervene and preserving the visual cues needed for 3D generation, while leaving clean inputs unchanged. We introduce Gate3DGen, a selective pre-processing model that gates image inputs to existing 3D generators without retraining them. A degradation-aware router bypasses inputs classified as clean and extracts features to guide purification of the remaining inputs. Conditioned on these features, a 3D-oriented one-step purifier purifies degraded inputs with spatial losses that encourage preservation of object identity, contours, and textural cues. To provide paired supervision, we construct a synthetic data engine that renders clean views of 3D assets and synthesises composite degradations. On the synthetic benchmark, the purifier improves image fidelity and downstream 3D geometry in only 2.3s on average, reducing TRELLIS.2 mesh distance by 61.5% relative to degraded inputs. VLM evaluation and human preferences on real photographs further support identity preservation and degradation correction. More demos and code are available at https://supra-3d.github.io/gate3dgen.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Object (grammar), Computer science, Artificial intelligence, Engineering, Computer vision, Fidelity

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange