Metraipolis › archive › dc_2a53497828fe

SCARF: A Scene-Adaptive Depth-Guided G-3DGS Encoder Accelerator with Semantic Reuse and Fused Dataflow — arXiv:2611.02365

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.02365
License (as archived): CC0-1.0
Archived: 2026-11-03T18:28:26.354Z
SHA-256 of the archived text: 6da71c93e8711f1605ab81fee92970da1a1fec18009a742e14a6e38296a2d785
Stable URL of this page: https://metraipolis.com/documento/dc_2a53497828fe

How to cite

SCARF: A Scene-Adaptive Depth-Guided G-3DGS Encoder Accelerator with Semantic Reuse and Fused Dataflow — arXiv:2611.02365 — arXiv (en), https://arxiv.org/abs/2611.02365 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2a53497828fe (Metraipolis archive dc_2a53497828fe, sha256 6da71c93e8711f1605ab81fee92970da1a1fec18009a742e14a6e38296a2d785)

BibTeX

@misc{dc2a53497828fe,
  title = {SCARF: A Scene-Adaptive Depth-Guided G-3DGS Encoder Accelerator with Semantic Reuse and Fused Dataflow — arXiv:2611.02365},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02365},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2a53497828fe; sha256 6da71c93e8711f1605ab81fee92970da1a1fec18009a742e14a6e38296a2d785},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02365. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Depth-guided, cost-volume-based Generalizable 3D Gaussian Splatting (G-3DGS) encoders reconstruct scenes as sets of Gaussian primitives directly from multi-view images without requiring per-scene optimization, thereby facilitating feed-forward novel-view synthesis on edge computing platforms. However, their shared encoder architecture presents two primary bottlenecks. First, the per-pixel depth search disregards semantic similarity within the 2D feature space, causing depth search operations to account for over 65% of memory traffic due to redundant exhaustive candidate scans. Second, the sequential pipeline of depth prediction followed by Gaussian generation neglects local 3D geometric continuity, resulting in more than 52% of generated Gaussians being redundant prior to instantiation. In this paper, we introduce SCARF, a dataflow-co-designed accelerator that aligns each bottleneck with a corresponding scene-level regularity. To mitigate the first bottleneck, Feature Similarity Depth Reuse reconceptualizes depth search as a problem of semantic locality: depth consistency is determined by affinity in feature space rather than physical address proximity. It transforms exhaustive per-pixel depth searches into anchor-centered local searches. To overcome the second bottleneck, Scene-Adaptive Early Sparsification disrupts the conventional serial depth-to-materialization boundary by probing each tile to infer local scene regularity. It then selectively activates only the minimally sufficient generation pathway, eliminating redundant Gaussian computations before primitive instantiation. On a 27.8 square millimeter 28 nm post-layout design, SCARF delivers a 2.94x geometric-mean end-to-end speedup with maximum PSNR degradation below 0.072%, 2.23 inferences per joule, and 34.5x GPU energy efficiency.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.AR, nuovi lavori del giorno).
Categorie arXiv: Hardware Architecture (cs.AR).
Pagina del lavoro: https://arxiv.org/abs/2611.02365 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T18:10:37.665Z dalla pagina https://arxiv.org/list/cs.AR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).