Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.10956 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911314491736064 |
|---|---|
| author | Zhou, Wentao Chen, Xuweiyi Rajagopal, Vignesh Chen, Jeffrey Chandra, Rohan Cheng, Zezhou |
| author_facet | Zhou, Wentao Chen, Xuweiyi Rajagopal, Vignesh Chen, Jeffrey Chandra, Rohan Cheng, Zezhou |
| contents | The success of foundation models in language and vision motivated research in fully end-to-end robot navigation foundation models (NFMs). NFMs directly map monocular visual input to control actions and ignore mid-level vision modules (tracking, depth estimation, etc) entirely. While the assumption that vision capabilities will emerge implicitly is compelling, it requires large amounts of pixel-to-action supervision that are difficult to obtain. The challenge is especially pronounced in dynamic and unstructured settings, where robust navigation requires precise geometric and dynamic understanding, while the depth-scale ambiguity in monocular views further limits accurate spatial reasoning. In this paper, we show that relying on monocular vision and ignoring mid-level vision priors is inefficient.
We present StereoWalker, which augments NFMs with stereo inputs and explicit mid-level vision such as depth estimation and dense pixel tracking. Our intuition is straightforward: stereo inputs resolve the depth-scale ambiguity, and modern mid-level vision models provide reliable geometric and motion structure in dynamic scenes. We also curate a large stereo navigation dataset with automatic action annotation from Internet stereo videos to support training of StereoWalker and to facilitate future research. Through our experiments, we find that mid-level vision enables StereoWalker to achieve a comparable performance as the state-of-the-art using only 1.5% of the training data, and surpasses the state-of-the-art using the full data. We also observe that stereo vision yields higher navigation performance than monocular input. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_10956 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Empowering Dynamic Urban Navigation with Stereo and Mid-Level Vision Zhou, Wentao Chen, Xuweiyi Rajagopal, Vignesh Chen, Jeffrey Chandra, Rohan Cheng, Zezhou Computer Vision and Pattern Recognition The success of foundation models in language and vision motivated research in fully end-to-end robot navigation foundation models (NFMs). NFMs directly map monocular visual input to control actions and ignore mid-level vision modules (tracking, depth estimation, etc) entirely. While the assumption that vision capabilities will emerge implicitly is compelling, it requires large amounts of pixel-to-action supervision that are difficult to obtain. The challenge is especially pronounced in dynamic and unstructured settings, where robust navigation requires precise geometric and dynamic understanding, while the depth-scale ambiguity in monocular views further limits accurate spatial reasoning. In this paper, we show that relying on monocular vision and ignoring mid-level vision priors is inefficient. We present StereoWalker, which augments NFMs with stereo inputs and explicit mid-level vision such as depth estimation and dense pixel tracking. Our intuition is straightforward: stereo inputs resolve the depth-scale ambiguity, and modern mid-level vision models provide reliable geometric and motion structure in dynamic scenes. We also curate a large stereo navigation dataset with automatic action annotation from Internet stereo videos to support training of StereoWalker and to facilitate future research. Through our experiments, we find that mid-level vision enables StereoWalker to achieve a comparable performance as the state-of-the-art using only 1.5% of the training data, and surpasses the state-of-the-art using the full data. We also observe that stereo vision yields higher navigation performance than monocular input. |
| title | Empowering Dynamic Urban Navigation with Stereo and Mid-Level Vision |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.10956 |