Michael Firman (@_mdfirman) header
Michael Firman (@_mdfirman) profile picture

Machine learning, computer vision, Python, trains, augmented reality etc

London
Joined October 2009
Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)Photo by Michael Firman (@_mdfirman)
I'm really excited to share our #CVPR2025 paper *MVSAnywhere*. This is a robust, sharp, and highly generalisable depth estimation system. This project was brilliantly led by @serizba: See his thread for all the beautiful videos, images and details 🧵👇
🔍Looking for a multi-view depth method that just works? We're excited to share MVSAnywhere, which we will present at #CVPR2025. MVSAnywhere produces sharp depths, generalizes and is robust to all kind of scenes, and it's scale agnostic. More info: nianticlabs.github.io/mvsany…
1
10
347
Take a look at this great splatting transformation work from my brilliant Niantic colleagues. It's really quite amazing. 👇🧵
Splats are great, but wouldn’t it be wonderful if we could transform them into the stuff of dreams? Introducing our #CVPR2025 paper ✨Morpheus✨! Morpheus lets you change the shape and appearance of 3D Gaussian splats with a few words! nianticlabs.github.io/morphe… (1/6)
6
169
Michael Firman retweeted
Splats are great, but wouldn’t it be wonderful if we could transform them into the stuff of dreams? Introducing our #CVPR2025 paper ✨Morpheus✨! Morpheus lets you change the shape and appearance of 3D Gaussian splats with a few words! nianticlabs.github.io/morphe… (1/6)
3
53
8
234
27,456
I am really excited to share this latest hand pose estimation work from Niantic, by @gui_ggh and @EugValassakis. This will appear at #ECCV2024 🎆 Take a look at the thread below 🧵👇
👋Would you like more accurate 3D camera-space hand mesh predictions? Check our new #ECCV2024 work: "HandDGP: Camera-Space Hand Mesh Prediction with Differentiable Global Positioning". Project page: nianticlabs.github.io/handdg… (1/5)
6
265
Michael Firman retweeted
Labour GAIN South East Cornwall from the Conservatives 🌹
14
48
24
379
81,041
🎉🎉👏👏💪💪 Congratulations @annagelderd MP!!
Thank you to the people of South East Cornwall for voting for a fresh start and putting your trust in me. We have made history with our first ever Labour MP here, and the hard work of change starts now.
1
78
If you happen to follow me and also live in S.E. Cornwall, please please head out and vote for @annagelderd on Thursday 🗳️🌹. Anna's dedication and passion for important causes are unmatched; South East Cornwall could not ask for a better advocate 💪. See the video for why 👇📽️
This election in South East Cornwall is about change. It’s our chance for our special corner of Cornwall to get its fair share from the next Government. From a Labour government. We need a fresh start.
6
344
Michael Firman retweeted
This election in South East Cornwall is about change. It’s our chance for our special corner of Cornwall to get its fair share from the next Government. From a Labour government. We need a fresh start.
28
61
12
202
95,323
I am really excited to share DoubleTake, our new approach for state-of-the-art depth and reconstruction. See @MohammedAmr1 's great thread 🧵👇 for details and shiny videos 🌟 – and see the project page for more! nianticlabs.github.io/double…
Are you frustrated of intermittently bad depth maps ruining your online pipeline 😫? Wish you could get fast 🚀 reuse of your previously predicted depth frames? Do you want shiny fast SOTA feedforward depth and meshes ✨? Introducing DoubleTake! nianticlabs.github.io/double…
1
13
736
Michael Firman retweeted
DoubleTake: Geometry Guided Depth Estimation @MohammedAmr1 @AleottiFilippo, Jamie Watson, Zawar Qureshi, @gui_ggh, Gabriel Brostow, Sara Vicente, @_mdfirman tl;dr: SimpleRecon + "hint MLP", processing cost volume + previous depth estimate. arxiv.org/abs/2406.18387
13
63
4,085
We are excited to present “AirPlanes ✈️: Accurate Plane Estimation via 3D-Consistent Embeddings” at #CVPR2024 in Seattle! AirPlanes ✈️ gives a lightweight and interpretable 3D planar reconstruction, for AR effects and more! 🤩 nianticlabs.github.io/airpla… How do we do this? 👇🧵
2
10
2
82
11,359
And if you are in Seattle ☕️🏔️⛴️ this week, please come by our poster at CVPR ✈️!
1
2
170
Our AirPlanes ✈️ poster is WED-PM: PS-2-42 Finally, please see @eric_brachmann's great thread of all of Niantic Research's work at #CVPR2024 🏆: teapawt.org/eric_brachmann/status/…
Catch Niantic Research at #CVPR2024! - I will give a talk at the MonoDepth challenge, TUE-PM. Teaser below! - Niantic presents two posters WED-PM: AirPlanes and MicKey. Do not miss the MicKey 📢Oral 📢 by @axelbarrosotw! - One more poster FRI-AM: marepo (✨Highlight✨)
1
4
626