Signals 4 · free daily AI digest

LangStreet: Persistent Language Fields for Anchor-Decoded Street Gaussians

Paper recorded by Signals 4 on 2026-09-10 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-10 on arXiv · recorded by Signals 4 on 2026-09-11

Category: cs.CV · 计算机视觉 · first seen 2026-09-11

Abstract

Language Gaussian fields implicitly assume that the primitive carrying semantics remains identifiable across views. This assumption breaks in scalable anchor-decoded representations, where persistent anchors generate view-conditioned child Gaussians whose geometry and appearance vary with the camera. We introduce Ours, a persistent language field for such structured Gaussian scenes. Our key idea i

Read on arXiv →

#90 most recent of 237 cs.CV papers we have recorded · ↑ newer: Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation · ↓ older: MMGait: Benchmarking and Unifying Gait Recognition across Heterogeneou
Cite this page: LangStreet: Persistent Language Fields for Anchor-Decoded Street Gaussians: the #90 most recent of 237 cs.CV papers we have recorded (as of 2026-09-10). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/langstreet-persistent-language-fields-for-anchor-decoded-street-gaussians.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CV papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions