Robot Odometry with a 4-Pixel sensor and an IMU.</p>\n","updatedAt":"2026-05-22T08:04:54.105Z","author":{"_id":"66f2e0cd5ddc45b044e955dd","avatarUrl":"/avatars/4df7f0be27282ac817ba747e6b025a1a.svg","fullname":"Francesco Pasti","name":"pastifra","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7384448647499084},"editors":["pastifra"],"editorAvatarUrls":["/avatars/4df7f0be27282ac817ba747e6b025a1a.svg"],"reactions":[],"isReport":false}},{"id":"6a1011ece317cbdad1e8c41d","author":{"_id":"66f2e0cd5ddc45b044e955dd","avatarUrl":"/avatars/4df7f0be27282ac817ba747e6b025a1a.svg","fullname":"Francesco Pasti","name":"pastifra","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false},"createdAt":"2026-05-22T08:21:00.000Z","type":"comment","data":{"edited":true,"hidden":true,"hiddenBy":"","hiddenReason":"Resolved","latest":{"raw":"This comment has been hidden","html":"This comment has been hidden","updatedAt":"2026-05-22T08:21:39.872Z","author":{"_id":"66f2e0cd5ddc45b044e955dd","avatarUrl":"/avatars/4df7f0be27282ac817ba747e6b025a1a.svg","fullname":"Francesco Pasti","name":"pastifra","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":1,"editors":[],"editorAvatarUrls":[],"reactions":[]}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.19990","authors":[{"_id":"6a0eb82b164dbbc68a26c75b","user":{"_id":"66f2e0cd5ddc45b044e955dd","avatarUrl":"/avatars/4df7f0be27282ac817ba747e6b025a1a.svg","isPro":false,"fullname":"Francesco Pasti","user":"pastifra","type":"user","name":"pastifra"},"name":"Francesco Pasti","status":"claimed_verified","statusLastChangedAt":"2026-05-21T19:21:01.228Z","hidden":false},{"_id":"6a0eb82b164dbbc68a26c75c","name":"Jeremy Klotz","hidden":false},{"_id":"6a0eb82b164dbbc68a26c75d","name":"Nicola Bellotto","hidden":false},{"_id":"6a0eb82b164dbbc68a26c75e","name":"Shree K. Nayar","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/66f2e0cd5ddc45b044e955dd/MuXNlgFzI17NK7KyG16-b.mp4"],"publishedAt":"2026-05-19T00:00:00.000Z","submittedOnDailyAt":"2026-05-22T00:00:00.000Z","title":"Minimalist Visual Inertial Odometry","submittedOnDailyBy":{"_id":"66f2e0cd5ddc45b044e955dd","avatarUrl":"/avatars/4df7f0be27282ac817ba747e6b025a1a.svg","isPro":false,"fullname":"Francesco Pasti","user":"pastifra","type":"user","name":"pastifra"},"summary":"Visual-Inertial Odometry(VIO), which is critical to mobile robot navigation, uses cameras with a large number of pixels. Capturing and processing camera images requires significant resources. This work presents a minimalist approach to planar odometry, demonstrating that just four visual measurements and an IMU can provide robust motion estimation for differential-drive robots. Our key insight is that four downward-facing photodiodes that sense the world through optical Gabor masks produce signals that encode speed. Based on this, we jointly optimize the mask parameters alongside a Temporal Convolutional Network (TCN) using a physically-grounded simulator. The resulting model decodes speed from just the four measurements produced by the photodiodes. Pairing these estimates with the angular speed from an IMU yields a continuous planar trajectory. We validate our approach with a prototype sensor mounted on a differential drive robot. Across diverse indoor and outdoor terrains, our system closely tracks the reference ground truth without any real-world fine-tuning. Our work shows that minimalist sensing enables efficient and accurate planar odometry.","upvotes":0,"discussionId":"6a0eb82c164dbbc68a26c75f","projectPage":"https://cave.cs.columbia.edu/projects/categories/project?cid=Computational%20Imaging&pid=Minimalist%20Visual%20Inertial%20Odometry","githubRepo":"https://github.com/pastifra/four-pixel-vio","githubRepoAddedBy":"user","ai_summary":"A minimalist visual-inertial odometry approach uses four photodiodes with optical Gabor masks and a temporal convolutional network to achieve accurate planar motion estimation for differential-drive robots.","ai_keywords":["Temporal Convolutional Network","visual-inertial odometry","photodiodes","optical Gabor masks","differential-drive robots","physically-grounded simulator"],"githubStars":4,"organization":{"_id":"63f68badb607296857bb2441","name":"columbia","fullname":"Columbia University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68e396f2b5bb631e9b2fac9a/USuDBWwfOGNrQ0SZGFxDF.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"organization":{"_id":"63f68badb607296857bb2441","name":"columbia","fullname":"Columbia University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68e396f2b5bb631e9b2fac9a/USuDBWwfOGNrQ0SZGFxDF.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.19990.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2605.19990 in a model README.md to link it from this page.

Cite arxiv.org/abs/2605.19990 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2605.19990 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.