Computer Science > Computer Vision and Pattern Recognition
[Submitted on 24 Feb 2024 (v1), last revised 30 Jun 2024 (this version, v7)]
Title:NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigation
View PDF HTML (experimental)Abstract:Vision-and-language navigation (VLN) stands as a key research problem of Embodied AI, aiming at enabling agents to navigate in unseen environments following linguistic instructions. In this field, generalization is a long-standing challenge, either to out-of-distribution scenes or from Sim to Real. In this paper, we propose NaVid, a video-based large vision language model (VLM), to mitigate such a generalization gap. NaVid makes the first endeavor to showcase the capability of VLMs to achieve state-of-the-art level navigation performance without any maps, odometers, or depth inputs. Following human instruction, NaVid only requires an on-the-fly video stream from a monocular RGB camera equipped on the robot to output the next-step action. Our formulation mimics how humans navigate and naturally gets rid of the problems introduced by odometer noises, and the Sim2Real gaps from map or depth inputs. Moreover, our video-based approach can effectively encode the historical observations of robots as spatio-temporal contexts for decision making and instruction following. We train NaVid with 510k navigation samples collected from continuous environments, including action-planning and instruction-reasoning samples, along with 763k large-scale web data. Extensive experiments show that NaVid achieves state-of-the-art performance in simulation environments and the real world, demonstrating superior cross-dataset and Sim2Real transfer. We thus believe our proposed VLM approach plans the next step for not only the navigation agents but also this research field.
Submission history
From: Jiazhao Zhang [view email][v1] Sat, 24 Feb 2024 16:39:16 UTC (43,696 KB)
[v2] Tue, 27 Feb 2024 17:00:03 UTC (43,696 KB)
[v3] Fri, 1 Mar 2024 05:09:02 UTC (43,696 KB)
[v4] Sat, 23 Mar 2024 16:54:01 UTC (43,696 KB)
[v5] Mon, 27 May 2024 08:47:03 UTC (44,371 KB)
[v6] Tue, 28 May 2024 04:48:19 UTC (44,371 KB)
[v7] Sun, 30 Jun 2024 11:14:13 UTC (44,393 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.