Marwan Taher

@marwan_ptr

PhD student at the Dyson Robotics Lab @ Imperial College London.

marwan99.github.io

Joined April 2011

514 Following

504 Followers

Replies Reposts Media

Pinned
Marwan Taher
@marwan_ptr
Dec 19, 2025
How can we run reconstruction models like π³ and Depth Anything 3 in real-time? We present KV-Tracker, a training-free approach, for real-time tracking of scenes and objects. Achieving up to 30 FPS! With @alzugarayign, @makezur, @XinKong_IC and @AjdDavison
11
Marwan Taher
@marwan_ptr
Jun 22
Download the code and give it a try!
Andrew Davison
@AjdDavison
Jun 22
Just announced: #CVPR2026 Best Demo Award for KV-Tracker. Congrats to Marwan and co-authors. And thanks to @deepfry_n, @shinjeong99 for massive help running the demo. That's 4 CVPR Best Demo Awards/Honourable Mentions in 3 years for our lab at Imperial! Proud of our demo culture.
Quote
Marwan Taher
@marwan_ptr
Dec 19, 2025
How can we run reconstruction models like π³ and Depth Anything 3 in real-time? We present KV-Tracker, a training-free approach, for real-time tracking of scenes and objects. Achieving up to 30 FPS! With @alzugarayign, @makezur, @XinKong_IC and @AjdDavison
1
Marwan Taher
@marwan_ptr
Jun 6
Wanna see a real-time demo of KV-Tracker? Stop by in ExHall F today 11:45 - 13:45! #CVPR2026
2
Marwan Taher
@marwan_ptr
Feb 26
KV-Tracker has been accepted to #CVPR2026!
Marwan Taher
@marwan_ptr
Dec 19, 2025
How can we run reconstruction models like π³ and Depth Anything 3 in real-time? We present KV-Tracker, a training-free approach, for real-time tracking of scenes and objects. Achieving up to 30 FPS! With @alzugarayign, @makezur, @XinKong_IC and @AjdDavison
2
Marwan Taher
@marwan_ptr
Dec 18, 2025
Per-frame geometry from π³ is split into primitives via segmentation and tracked over time using dense 2D point tracks. With a compact per primitive pose, geometry is densely aligned, stitching primitives to create a complete reconstruction of the observed scene components.
Kirill Mazur
@makezur
Dec 18, 2025
Introducing 4D Primitive-Mâché (4DPM), a new method for replayable 4D reconstruction from monocular videos. We split dynamic scenes into 3D primitives and recover their motion. 4DPM can infer object positions even after they leave view. Joint work with @marwan_ptr @AjdDavison

Marwan Taher

Marwan Taher’s posts