Simultaneous localization and mapping (SLAM) is one of the fundamental problems in robotics, as it enables autonomous operations in real-world scenarios. Under low illumination, reduced contrast, sensor noise, and motion blur degrade both feature extraction and feature matching, while compensating with LiDAR, depth, or thermal sensors raise cost, power draw, and integration complexity. Existing benchmarks remain dominated by well-lit indoor or daylight sequences, leaving open how far SLAM with standard RGB cameras can be pushed in the dark. We benchmark six systems spanning feature-based, direct, filter-based, and learning-based paradigms—ORB-SLAM3, DSO, Kimera-VIO, OpenVINS, DPVO, and DPV-SLAM—on five LaMARia sequences of varying difficulty and illumination, reporting absolute and relative pose error alongside control-point recall. Kimera-VIO is the only system to track all five sequences to completion, combining the lowest relative pose error with steadily growing absolute error due to the absence of loop closure; DPVO and DPV SLAM never lose tracking but incur absolute errors of roughly 100 m under low light; and the classical monocular pipelines (ORB-SLAM3, DSO) together with the filter-based OpenVINS fail outright or diverge on most of the harder and low-light sequences. The results suggest that RGB-only SLAM maintains stable low-light tracking only when both inertial fusion and global optimization are present. Closing the remaining gap will likely require low-light-specific learned front-ends or a return to complementary sensing.</p>\n","updatedAt":"2026-07-23T13:53:12.243Z","author":{"_id":"662798e43e2a8eec1b53ed1a","avatarUrl":"/avatars/28871a0725cf49347457fa8535ce5fdb.svg","fullname":"Yaroslav Prytula","name":"YaroslavPrytula","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":2,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8855780959129333},"editors":["YaroslavPrytula"],"editorAvatarUrls":["/avatars/28871a0725cf49347457fa8535ce5fdb.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.17699","authors":[{"_id":"6a6219ff2ee212ed0e2a1224","name":"Oleh Basystyi","hidden":false},{"_id":"6a6219ff2ee212ed0e2a1225","name":"Anna Stasyshyn","hidden":false},{"_id":"6a6219ff2ee212ed0e2a1226","name":"Oleksandr Kosovan","hidden":false},{"_id":"6a6219ff2ee212ed0e2a1227","name":"Yaroslav Prytula","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/662798e43e2a8eec1b53ed1a/bdwKoA-I341k6vXC6pfQ0.jpeg","https://cdn-uploads.huggingface.co/production/uploads/662798e43e2a8eec1b53ed1a/N8vJIzxjHmRlsytOGUErX.jpeg","https://cdn-uploads.huggingface.co/production/uploads/662798e43e2a8eec1b53ed1a/eRbRwvbIMHtPCFVUmySmu.jpeg"],"publishedAt":"2026-07-20T00:00:00.000Z","submittedOnDailyAt":"2026-07-23T00:00:00.000Z","title":"SLAM in Low-Light Environments: Project Report","submittedOnDailyBy":{"_id":"662798e43e2a8eec1b53ed1a","avatarUrl":"/avatars/28871a0725cf49347457fa8535ce5fdb.svg","isPro":false,"fullname":"Yaroslav Prytula","user":"YaroslavPrytula","type":"user","name":"YaroslavPrytula"},"summary":"Simultaneous localization and mapping (SLAM) is one of the fundamental problems in robotics, as it enables autonomous operations in real-world scenarios. Under low illumination, reduced contrast, sensor noise, and motion blur degrade both feature extraction and feature matching, while compensating with LiDAR, depth, or thermal sensors raises cost, power draw, and integration complexity. Existing benchmarks remain dominated by well-lit indoor or daylight sequences, leaving open how far SLAM with standard RGB cameras can be pushed in the dark. We benchmark six systems spanning the feature-based, direct, filter-based, and learning-based paradigms - ORB-SLAM3, DSO, Kimera-VIO, OpenVINS, DPVO, and DPV-SLAM - on five LaMARia sequences of varying difficulty and illumination, reporting absolute and relative pose error alongside control-point recall. Kimera-VIO is the only system to track all five sequences to completion, combining the lowest relative pose error with steadily growing absolute error due to the absence of loop closure; DPVO and DPV-SLAM never lose tracking but incur absolute errors of roughly 100 m under low light; and the classical monocular pipelines (ORB-SLAM3, DSO) together with the filter-based OpenVINS fail outright or diverge on most of the harder and low-light sequences. The results suggest that RGB-only SLAM maintains stable low-light tracking only when both inertial fusion and global optimization are present. Closing the remaining gap will likely require low-light-specific learned front-ends or a return to complementary sensing.","upvotes":0,"discussionId":"6a6219ff2ee212ed0e2a1228","projectPage":"https://ucu-autonomous-ugv.github.io/research/","organization":{"_id":"6a3bde9e0820652bfdaa306a","name":"ucu-autonomous-ugv","fullname":"UCU Autonomous UGV Club","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/654cd3283a321f06c2b66521/wYOG0O1hhF2j-BKHfdpYh.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"organization":{"_id":"6a3bde9e0820652bfdaa306a","name":"ucu-autonomous-ugv","fullname":"UCU Autonomous UGV Club","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/654cd3283a321f06c2b66521/wYOG0O1hhF2j-BKHfdpYh.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.17699.md","query":{}}">
SLAM in Low-Light Environments: Project Report
Abstract
Simultaneous localization and mapping (SLAM) is one of the fundamental problems in robotics, as it enables autonomous operations in real-world scenarios. Under low illumination, reduced contrast, sensor noise, and motion blur degrade both feature extraction and feature matching, while compensating with LiDAR, depth, or thermal sensors raises cost, power draw, and integration complexity. Existing benchmarks remain dominated by well-lit indoor or daylight sequences, leaving open how far SLAM with standard RGB cameras can be pushed in the dark. We benchmark six systems spanning the feature-based, direct, filter-based, and learning-based paradigms - ORB-SLAM3, DSO, Kimera-VIO, OpenVINS, DPVO, and DPV-SLAM - on five LaMARia sequences of varying difficulty and illumination, reporting absolute and relative pose error alongside control-point recall. Kimera-VIO is the only system to track all five sequences to completion, combining the lowest relative pose error with steadily growing absolute error due to the absence of loop closure; DPVO and DPV-SLAM never lose tracking but incur absolute errors of roughly 100 m under low light; and the classical monocular pipelines (ORB-SLAM3, DSO) together with the filter-based OpenVINS fail outright or diverge on most of the harder and low-light sequences. The results suggest that RGB-only SLAM maintains stable low-light tracking only when both inertial fusion and global optimization are present. Closing the remaining gap will likely require low-light-specific learned front-ends or a return to complementary sensing.
Community
Simultaneous localization and mapping (SLAM) is one of the fundamental problems in robotics, as it enables autonomous operations in real-world scenarios. Under low illumination, reduced contrast, sensor noise, and motion blur degrade both feature extraction and feature matching, while compensating with LiDAR, depth, or thermal sensors raise cost, power draw, and integration complexity. Existing benchmarks remain dominated by well-lit indoor or daylight sequences, leaving open how far SLAM with standard RGB cameras can be pushed in the dark. We benchmark six systems spanning feature-based, direct, filter-based, and learning-based paradigms—ORB-SLAM3, DSO, Kimera-VIO, OpenVINS, DPVO, and DPV-SLAM—on five LaMARia sequences of varying difficulty and illumination, reporting absolute and relative pose error alongside control-point recall. Kimera-VIO is the only system to track all five sequences to completion, combining the lowest relative pose error with steadily growing absolute error due to the absence of loop closure; DPVO and DPV SLAM never lose tracking but incur absolute errors of roughly 100 m under low light; and the classical monocular pipelines (ORB-SLAM3, DSO) together with the filter-based OpenVINS fail outright or diverge on most of the harder and low-light sequences. The results suggest that RGB-only SLAM maintains stable low-light tracking only when both inertial fusion and global optimization are present. Closing the remaining gap will likely require low-light-specific learned front-ends or a return to complementary sensing.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2607.17699 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.17699 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.17699 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.