🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Excited about this #ICRA2020 paper with Tristan Laidlow, and others: we train a CNN to predict probability distributions for the depth of each pixel and fuse this with estimates from a multi-view stereo algorithm to improve dense 3D reconstructions. doc.ic.ac.uk/~sleutene/publ…

Excited about this #ICRA2020 paper with Anna Dai, Sotiris Papatheodorou, and others: arxiv.org/pdf/2002.04440 We use high-res octree-based probabilistic occupancy mapping within a planning and frontiers for fast exploration of unknown space. youtu.be/tH2VkVony38 via YouTube

Excited about this ICRA2020 paper with Dimos Tzoumanikas and Cheryl Yan: arxiv.org/pdf/2002.06598 We use the full non-linear drone dynamics to perform aggressive manoeuvres using NMPC, while identifying and recovering from motor failures. youtu.be/cAQeSZ3tIqY via YouTube

Deep Learning and SLAM Research Intern - SLAMcore careers.slamcore.com/jobs/874411-de… via Slamcore

Excited about this #RSS2020 paper with Dimos Tzoumanikas, Felix Graule, Cheryl Yan, Dhruv Shah and Marija Popovic. We propose a hybrid force and position NMPC for aerial manipulation showcased by a writing drone. arxiv.org/pdf/2006.02116…, youtu.be/iE--MO0YF0o via YouTube

Thanks for pulling this together, Guillermo Gallego and all!

At #EVS20 Owen Nicholson demos #SpatialAI SDK as commercial grade #VisualSLAM solution to help #Robots get from A to B safely and reliably. Download in 5 clicks and plug&play compatible with range of sensors incl. Intel RealSense. Watch the demo here hubs.ly/H0w9sWJ0

Quick personal & professional update: per 1st February '21, I will be starting at TU Munich (TT Prof). Imperial was great, but good to be back in (continental) Europe! Links with Slamcore unchanged. Onwards to more great research & looking for a few great researchers to join!

In this RA-L/ICRA’21 work with Binbin Xu and Andy Davison, we investigate uncertainty/weighting of learned dense features for camera/obj. tracking in drastic lighting changes, trained end-to-end with pose prediction for robust initialisation. youtu.be/6pMosl6ZAPE via YouTube

In this RA-L/ICRA work (Funk​, Tarrio Slamcore, Papatheodorou, Popović​, Alcantarilla Slamcore, and me) we adopt adaptive-resolution into fast&accurate occupancy mapping for fast collision queries in planning. arxiv.org/pdf/2010.07929… youtu.be/XlnO3GBNPvc via YouTube

Thanks everyone for your patience. The BMVA and PCs are able to announce a timeline for the virtual BMVC 2021 (deadline 25 June, conference 22-25 Nov): …ishmachinevisionassociation.github.io/bmvc Also, excited to announce a new BMVA in-person event in Sept in Manchester: …ishmachinevisionassociation.github.io/meetings/21-09…

🥳 Our work "Volumetric Occupancy Mapping with Probabilistic Depth Completion for Robotic Navigation" will appear soon in IEEE RA-L! Co-authors: Florian Thomas, Sotiris Papatheodorou, Nils Funk, Teresa Vidal & Stefan Leutenegger. arxiv.org/abs/2012.03023 youtube.com/watch?v=7n0I8S…

Yiduo's #ICRA2021 paper is focused on large-scale reconstruction with LiDAR. A great collaboration with Imperial College (now in TUM and Bonn!) Video: youtu.be/-_vtsAyHXOg Paper: robots.ox.ac.uk/~mobile/drs/Pa… Session: Wed, Jun 2, 9:30 (physical session)

📢 Keynotes Announcement 📢 Exciting news! We have 4 fantastic keynotes lined up: - Prof. Andrew Zisserman (Visual Geometry Group (VGG)) - Prof. Daphne Koller (@daphnekoller) - Prof. Davide Scaramuzza (Davide Scaramuzza) - Prof. Katerina Fragkiadaki bmvc2021.com/programme/keyn…

Today we are kicking off #BMVC2021 Less than one hour to go until the first Poster Session of the conference! Check out the exciting schedule for today with a keynote talk from Prof. Andrew Zisserman on "How can we learn sign language by watching sign-interpreted TV?"

Advancing commercialization of Spatial AI for robots, today we announce Series A funding from @ROBOGlobal, Presidio Ventures, Amadeus Capital Partners, Global Brain, IP Group - visionary ventures MMC Octopus Ventures, Samsung Ventures, Toyota Ventures ヤマト運輸 hubs.ly/Q019n1Dw0

In our #ECCV2022 paper/poster (Henning, Laidlow, and me), we show that a joint optimisation of camera and human poses, shape, and postures delivers accurate results with everything in motion -- including metric scale. Paper: arxiv.org/abs/2205.02301 Video: youtu.be/0-SL3VeWEvU

In our #IROS2022 paper, we (Binbin, Andrew Davison, and me) introduce a way of object-level mapping that conditions on the depth-fused volume and a shape prior. It preserves observed details and otherwise completes for better tracking & mapping. youtu.be/mH22H7jp1D8 via YouTube

In our ICRA'23 work (S. Papatheodorou, N. Funk, D. Tzoumanikas, C. Choi, Binbin Xu, me), we broaden the scope of exploration beyond free space: an MAV finds specific objects in the unknown and reconstructs them to a target level of detail. youtu.be/z0LVe_8SATU via YouTube

Excited to present DiffCAD coming to #SIGGRAPH2024! Daoyi Gao introduces the first probabilistic single-view CAD retrieval & alignment. We train only on synthetic -> generalize robustly to real images! Check out the code: daoyig.github.io/DiffCAD_/ w/David Rozenberszki, Stefan Leutenegger