Newest papers

CoGaussian: Open-Vocabulary Gaussian-based Collaborative 3D Scene Understanding

Om Shivam Verma, Shaya Farahmand, Lingjun Zhao, Lu Gan · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28530 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

A percepção colaborativa permite que os agentes recuperem a semântica e a geometria da cena além do campo de visão de qualquer agente individual. Esta compreensão mais ampla da cena aumenta a segurança de transporte e melhora a eficiência em sistemas multi-robot. No entanto, os métodos de ocupação semântica colaborativa existentes dependem de anotações de voxel 3D densas e estão limitados a reconhecer objetos dentro de um conjunto pré-definido de classes semânticas. Para superar essas limitações, apresentamos o CoGaussian, um framework de previsão de ocupação semântica colaborativa baseado em vocabulário aberto baseado em Gaussian. O CoGaussian representa uma cena como um conjunto de modelos de base alinhados com 3D Gaussians, que serve como uma compactação para comunicar e representar informações geométricas

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

Abstract

Collaborative perception enables agents to retrieve the scene semantics and geometry beyond the field of view of any individual agent. This broader scene understanding enhances transportation safety and improves efficiency in multi-robot systems. However, existing collaborative semantic occupancy methods rely on dense 3D voxel annotations and are limited to recognizing objects within a predefined set of semantic classes. To overcome these limitations, we present CoGaussian, an open-vocabulary Gaussian-based collaborative semantic occupancy prediction framework. CoGaussian represents a scene as a set of foundation model-aligned sparse 3D Gaussians, which serve as a compact representation for communicating and fusing geometric and semantic information across multiple agents. On the Semantic-OPV2V benchmark, our method outperforms existing open-vocabulary single-agent approaches, highlighting the effectiveness of open-vocabulary collaborative scene understanding.

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

Topics: Computer science, Semantics (computer science), Set (abstract data type), Representation (politics), Field (mathematics), Artificial intelligence

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