Newest papers

KoalaGS: Complexity-Aware Continuous Level of Detail for Gaussian Splatting

Radu Beche, Raoul de Charette · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28615 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Mengimplementasikan adegan 3D Gaussian Splatting yang terlatih di bawah pengurangan anggaran komputasi membutuhkan satu set primitif untuk tetap efektif ketika ditampilkan pada prefiks yang lebih kecil. Dalam pengaturan tingkat detail (LoD) yang berkelanjutan ini, tantangan adalah untuk memutuskan mana Gaussians harus mempertahankan kapasitas saat anggaran menurun. Sebuah studi pilot pada rekonstruksi 3DGS-MCMC menunjukkan bahwa kompleksitas gambar, yang dihitung dari pandangan input sebelum optimalisasi, memprediksi di mana kesalahan rekonstruksi terkonsentrasi dalam gambar dan ruang 3D. Hal ini juga mengungkapkan bahwa wilayah sederhana mengandung terlalu banyak Gaussians dan merosot sebanyak yang kompleks pada anggaran yang lebih rendah, meskipun secara inheren lebih murah untuk mewakili. Selanjutnya, di KoalaGS, selama pelatihan kami menggunakan kompleksitas ini untuk prioritas bias relokasi ke daerah

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

Abstract

Deploying a trained 3D Gaussian Splatting scene under shrinking compute budgets requires one primitive set to remain effective when rendered at smaller prefixes. In this continuous level of detail (LoD) setting, the challenge is to decide which Gaussians should retain capacity as the budget decreases. A pilot study on 3DGS-MCMC reconstructions shows that image complexity, computed from the input views before optimization, predicts where reconstruction error concentrates in image and 3D space. It also reveals that simple regions contain too many Gaussians and degrade as much as complex ones at lower budgets, despite being inherently cheaper to represent. Subsequently, in KoalaGS, during training we use this complexity prior to bias relocation toward poorly reconstructed areas while ensuring smooth degradation across LoDs. A primitive importance score then enables inference at low-capacity, while preserving the scene structure. Tested across three datasets, KoalaGS maintains competitive full-capacity quality while significantly improving quality across the budget spectrum, yielding smoother transitions across LoDs, and enabling faster progressive streaming.

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

Topics: Computer science, Artificial intelligence, Gaussian, Inference, Set (abstract data type), Computer vision

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