
@fdellaerti
iAccount based inUnited States
About this account
- Account based in
- United States
- Connected via
- United States App Store
Account-level information from X, not a live location or the device used for a specific post.
Robotics & Computer Vision Professor at Georgia Tech, and part-time CAIO at Verdant Robotics. Before: stints at KUL, Skydio, Facebook B*8, Google AI.
San Mateo, CA
Joined June 2008
- Tweets3.6K
- Following1.5K
- Followers17K
- Likes15.3K
Pinned Tweet
My Annual Reviews article on Factor Graphs in Robotics is finally out with a publicly accessible link: annualreviews.org/eprint/85P…
Frank Dellaert retweeted
FoundationVIO: Enforcing Metric Scale in Feed-forward Visual-Inertial Odometry
Jin Rhee, Yuhao Zhang, @fdellaert, @MauriceFallon
tl;dr: IMU measurements + GFM predictions->factor graph
arxiv.org/abs/2610.14951
Did I say GTSAM 4.3 is available? If using @gtsam4 I recommend switching to the latest version. We'll keep develop on a 4.3 version for now to quickly react to any breaking changes - but really only expect better performance :-) gtsam.org/
Frank Dellaert retweeted
Meet prototype 1 of the Goose, the family robot.
We're building for the home: tidy, play, live.
Frank Dellaert retweeted
😊🙏🐝💪
• #1 Public Best Value College Overall: Georgia Tech
• #1 Private Best Value College Overall: Princeton
princetonreview.com/press/be…
Frank Dellaert retweeted
🎉 Excited to share our recent work on deformable object manipulation, 𝗣𝗵𝘆𝘀𝗖𝗼𝗥𝗲, accepted to 𝗖𝗼𝗥𝗟 𝟮𝟬𝟮𝟲. Congrats to @hcy1n @ShuohanT!
𝗖𝗼𝗱𝗲 & 𝗗𝗮𝘁𝗮: lunarlab-gatech.github.io/Ph…
#CoRL2026 #RobotLearning #DeformableObjects
Excited to share our #CoRL2026 work PhysCoRe: a physics-corrected world model for deformable objects!
We couple a physics-based simulator with two feed-forward modules: one infers material from vision, the other corrects its internal dynamics.
Check: lunarlab-gatech.github.io/Ph…
So proud of my daughter and her PlosOne publication on gene expression in corals, when they are under thermal stress.
journals.plos.org/plosone/ar…
You fine-tune a robot foundation model on a hard task. It gets 25%. Now what?
Introducing Q-Planning, a learning-based harness that lets large black-box robot policies recursively self-improve.
On a hard fine-grained task: 25% → 80% in 100 robot attempts (~30 mins), no extra human data.
q-planning.github.io/
1/n 🧵
GTSAM is going GPU! 🚀
Ruogu Li has built an impressive new experimental CUDA backend for nonlinear factor-graph optimization in GTSAM, just in time for the 4.3 release next week.
Read our blog post:
gtsam.org/2026/08/20/cuda-ba…
#GTSAM #CUDA #Robotics #SLAM #StructureFromMotion #BundleAdjustment #ComputerVision #GPU #OpenSource
🤖 Made with AI
Follow up PR also made SFM much easier to use, see borglab.github.io/gtsam/sfm/
Exciting to see how the community puts @gtsam4 to work! In a guest post, Jash Shah explores an application of factor graphs to 3D Gaussian Splatting, bringing neural rendering together with pose graphs, loop closures, robust noise models, and iSAM2.
gtsam.org/2026/08/04/gaussia…
#GTSAM #SLAM #3DGS





















