Newest papers

Learning from Automatic 3D Gaussian Splatting Edits for Online Bayesian Scene Change Detection

Changrui Zhu, Simon Julier · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28490 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Deteksi perubahan adegan online membutuhkan integrasi pengamatan masuk untuk menentukan daerah mana yang telah berubah dibandingkan dengan adegan referensi. Kami memperbarui keyakinan perubahan saat gambar baru tiba menggunakan fusi bukti Bayesian umum. Namun, probabilitas awal perubahan dan model bukti tidak diketahui sebelumnya, dan inferensi Bayesian numerik dapat mahal. Kami mengatasi tantangan ini dengan belajar secara offline dari edisi otomatis ke model referensi 3D Gaussian Splatting (3DGS). Edisi ini menghasilkan tampilan pelatihan yang ditandai tanpa anotasi manual. Bersama dengan pengamatan referensi yang tidak berubah, mereka memberikan pengawasan untuk model bukti spesifik adegan belajar dan memperkirakan statistik piksel yang efektif dan tanggapan. Prediksi online menggabungkan bukti saat ini dengan dukungan historis, sementara faktor lokal Bayes berbentuk tertutup, diparameterisasi oleh statistik tanggapan yang dipelajari,

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

Abstract

Online scene change detection requires integrating incoming observations to determine which regions have changed relative to a reference scene. We update change beliefs as new images arrive using generalised Bayesian evidence fusion. However, the prior probability of change and evidence models are unknown beforehand, and numerical Bayesian inference can be costly. We address these challenges by learning offline from automatic edits to a reference 3D Gaussian Splatting (3DGS) model. These edits generate labelled training views without manual annotations. Together with unchanged reference observations, they provide supervision for learning scene-specific evidence models and estimating the effective pixel prior and response statistics. Online prediction combines current evidence with historical support, while closed-form local Bayes factors, parameterized by the learned response statistics, update an accumulated 3D Gaussian field. Object-guided refinement produces the final change masks. Our method achieves state-of-the-art accuracy on PASLCD and CCS3D, with $67.64\%$ and $75.82\%$ mIoU, respectively. On PASLCD, the complete pipeline runs at $17.2$\,FPS over complete query sequences, the fastest among the evaluated online methods.

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

Topics: Computer science, Change detection, Artificial intelligence, Bayesian probability, Inference, Gaussian process

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