Unifying 2D and 3D Vision-Language Understanding

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jain, Ayush, Swerdlow, Alexander, Wang, Yuzhou, Arnaud, Sergio, Martin, Ada, Sax, Alexander, Meier, Franziska, Fragkiadaki, Katerina
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909642171351040
author Jain, Ayush
Swerdlow, Alexander
Wang, Yuzhou
Arnaud, Sergio
Martin, Ada
Sax, Alexander
Meier, Franziska
Fragkiadaki, Katerina
author_facet Jain, Ayush
Swerdlow, Alexander
Wang, Yuzhou
Arnaud, Sergio
Martin, Ada
Sax, Alexander
Meier, Franziska
Fragkiadaki, Katerina
contents Progress in 3D vision-language learning has been hindered by the scarcity of large-scale 3D datasets. We introduce UniVLG, a unified architecture for 2D and 3D vision-language understanding that bridges the gap between existing 2D-centric models and the rich 3D sensory data available in embodied systems. Our approach initializes most model weights from pre-trained 2D models and trains on both 2D and 3D vision-language data. We propose a novel language-conditioned mask decoder shared across 2D and 3D modalities to ground objects effectively in both RGB and RGB-D images, outperforming box-based approaches. To further reduce the domain gap between 2D and 3D, we incorporate 2D-to-3D lifting strategies, enabling UniVLG to utilize 2D data to enhance 3D performance. With these innovations, our model achieves state-of-the-art performance across multiple 3D vision-language grounding tasks, demonstrating the potential of transferring advances from 2D vision-language learning to the data-constrained 3D domain. Furthermore, co-training on both 2D and 3D data enhances performance across modalities without sacrificing 2D capabilities. By removing the reliance on 3D mesh reconstruction and ground-truth object proposals, UniVLG sets a new standard for realistic, embodied-aligned evaluation. Code and additional visualizations are available at https://univlg.github.io .
format Preprint
id arxiv_https___arxiv_org_abs_2503_10745
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unifying 2D and 3D Vision-Language Understanding
Jain, Ayush
Swerdlow, Alexander
Wang, Yuzhou
Arnaud, Sergio
Martin, Ada
Sax, Alexander
Meier, Franziska
Fragkiadaki, Katerina
Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
Progress in 3D vision-language learning has been hindered by the scarcity of large-scale 3D datasets. We introduce UniVLG, a unified architecture for 2D and 3D vision-language understanding that bridges the gap between existing 2D-centric models and the rich 3D sensory data available in embodied systems. Our approach initializes most model weights from pre-trained 2D models and trains on both 2D and 3D vision-language data. We propose a novel language-conditioned mask decoder shared across 2D and 3D modalities to ground objects effectively in both RGB and RGB-D images, outperforming box-based approaches. To further reduce the domain gap between 2D and 3D, we incorporate 2D-to-3D lifting strategies, enabling UniVLG to utilize 2D data to enhance 3D performance. With these innovations, our model achieves state-of-the-art performance across multiple 3D vision-language grounding tasks, demonstrating the potential of transferring advances from 2D vision-language learning to the data-constrained 3D domain. Furthermore, co-training on both 2D and 3D data enhances performance across modalities without sacrificing 2D capabilities. By removing the reliance on 3D mesh reconstruction and ground-truth object proposals, UniVLG sets a new standard for realistic, embodied-aligned evaluation. Code and additional visualizations are available at https://univlg.github.io .
title Unifying 2D and 3D Vision-Language Understanding
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
url https://arxiv.org/abs/2503.10745