Links

• Dalù Decò website

News About Me

pySLAM v.2.10.0 – A Hybrid Python/C++ Framework for Visual SLAM and 3D Perception
RSS 2025 Workshop: Unifying Visual SLAM – From Fragmented Datasets to Scalable, Real-World Solutions
A new release of pySLAM is here. Loop-closing, volumetric integration, depth prediction and more improvements.
pySLAM v2.1 is out
PLVS code released

Blog

DreamDojo: Scaling Robot World Models with 44,000+ Hours of Egocentric Human Video
Probing the 3D Awareness of Visual Foundation Models
Waymo has unveiled the Waymo World Model
Figure AI Announces Helix 02 – a General-Purpose Humanoid System
Genie 3 – AI That Builds Worlds from Words

Tags

2D to 3D 3D reconstruction 3d scene understanding augmented reality business CNN computer vision data analysis dataset deep learning disaster robotics drones energy features Foundation models generative AI gps image processig inertial interactive AI lidar machine learning mapping math multi-robot NN open source perception place recognition representation learning robotics self-driving car sensor-based motion planning sensors sim-to-real simulation SLAM TRADR UGVs USAR video generation visual localization visual servoing VR world models

Tweets

Retweet on Twitter Luigi Freda Retweeted

Haian Jin @haian_jin ·

5 Mar

Spatial reconstruction is a long-context problem: real scenes come with hundreds of images. But O(N²) transformer-based models don’t scale efficiently.

Introducing: 🤐ZipMap (CVPR ’26): Linear-Time, Stateful 3D Reconstruction via Test-Time Training (TTT).

ZipMap “zips” a large

Reply on Twitter 2029676163476451769 Retweet on Twitter 2029676163476451769 98 Like on Twitter 2029676163476451769 731 Twitter 2029676163476451769

Retweet on Twitter Luigi Freda Retweeted

Shubham Tulsiani @shubhtuls ·

27 Feb

[1/N] Current visual geometry prediction models primarily rely on labeled 3D data. Our CVPR26 paper, Flow3r, allows additionally leveraging unlabeled videos (using flow supervision) for scalable visual geometry learning, enabling accurate multi-view 3D reconstruction in-the-wild.

Reply on Twitter 2027441878958755968 Retweet on Twitter 2027441878958755968 27 Like on Twitter 2027441878958755968 207 Twitter 2027441878958755968