Kirill Mazur Profile
Kirill Mazur

@makezur

Followers
346
Following
867
Media
5
Statuses
18

PhD student at Dyson Robotics Lab @ Imperial College London

London, UK
Joined May 2020
Don't wanna be here? Send us removal request.
@makezur
Kirill Mazur
2 years
SuperPrimitives enable 3D scene reconstruction at a level of local image regions rather than pixels. With SuperPrimitves, 3D reconstruction is done simply, thanks to powerful off-the-shelf 2D priors. project page: https://t.co/l360soWiAz video: https://t.co/H3GJ81qxx2
3
46
190
@rmurai0610
Riku Murai
11 months
Introducing MASt3R-SLAM, the first real-time monocular dense SLAM with MASt3R as a foundation. Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map. With @eric_dexheimer*, @AjdDavison (*Equal Contribution)
42
260
1K
@makezur
Kirill Mazur
1 year
@makezur
Kirill Mazur
1 year
Come by our poster this afternoon! ๐Ÿญ๐Ÿณ:๐Ÿญ๐Ÿฑ ๐˜๐—ผ ๐Ÿญ๐Ÿด:๐Ÿฐ๐Ÿฑ, ๐—”๐—ฟ๐—ฐ๐—ต ๐Ÿฐ๐—”-๐—˜
0
0
2
@makezur
Kirill Mazur
1 year
Curious how DSINE surface normals can be converted into full 3D geometry? Visit our demo of ๐—ฆ๐˜‚๐—ฝ๐—ฒ๐—ฟ๐—ฃ๐—ฟ๐—ถ๐—บ๐—ถ๐˜๐—ถ๐˜ƒ๐—ฒ-based real-time 3D reconstruction tomorrow (Friday)! A demo is worth a thousand pictures ๐Ÿ™‚ Demo: 10:30 to 18:45, Arch CDE #9 #CVPR #CVPR2024
@makezur
Kirill Mazur
1 year
SuperPrimitives will be presented at #CVPR next week (Wednesday), along with a ๐—ฟ๐—ฒ๐—ฎ๐—น-๐˜๐—ถ๐—บ๐—ฒ ๐—ฑ๐—ฒ๐—บ๐—ผ on Friday! Our new representation enables dense monocular 3D reconstruction in real-time. No poses required! Project page: https://t.co/l360soWiAz
0
1
14
@makezur
Kirill Mazur
1 year
Poster 14!
0
0
5
@makezur
Kirill Mazur
1 year
Come by our poster this afternoon! ๐Ÿญ๐Ÿณ:๐Ÿญ๐Ÿฑ ๐˜๐—ผ ๐Ÿญ๐Ÿด:๐Ÿฐ๐Ÿฑ, ๐—”๐—ฟ๐—ฐ๐—ต ๐Ÿฐ๐—”-๐—˜
@makezur
Kirill Mazur
1 year
SuperPrimitives will be presented at #CVPR next week (Wednesday), along with a ๐—ฟ๐—ฒ๐—ฎ๐—น-๐˜๐—ถ๐—บ๐—ฒ ๐—ฑ๐—ฒ๐—บ๐—ผ on Friday! Our new representation enables dense monocular 3D reconstruction in real-time. No poses required! Project page: https://t.co/l360soWiAz
3
9
51
@makezur
Kirill Mazur
1 year
SuperPrimitives will be presented at #CVPR next week (Wednesday), along with a ๐—ฟ๐—ฒ๐—ฎ๐—น-๐˜๐—ถ๐—บ๐—ฒ ๐—ฑ๐—ฒ๐—บ๐—ผ on Friday! Our new representation enables dense monocular 3D reconstruction in real-time. No poses required! Project page: https://t.co/l360soWiAz
3
43
213
@eric_dexheimer
Eric Dexheimer
2 years
How can we infer 3D-consistent poses and dense geometry in real-time given only RGB images? ๐—–๐—ข๐— ๐—ข decodes dense geometry from a compact and optimizable set of 3D anchor points to enforce 3D consistency. Project page: https://t.co/57wOkz6McV Work with @AjdDavison 1/n
4
56
267
@makezur
Kirill Mazur
2 years
Code release for SuperPrimitives, and it comes with an interactive GUI! #CVPR2024 https://t.co/ZWGwYhbVCF SuperPrimitive is a new 3D representation which enables solving many 3D tasks at the level of image segments.
1
38
172
@makezur
Kirill Mazur
2 years
SuperPrimitives got accepted to #CVPR2024! The code will be released soon and see you all in Seattle!
@makezur
Kirill Mazur
2 years
SuperPrimitives enable 3D scene reconstruction at a level of local image regions rather than pixels. With SuperPrimitves, 3D reconstruction is done simply, thanks to powerful off-the-shelf 2D priors. project page: https://t.co/l360soWiAz video: https://t.co/H3GJ81qxx2
2
7
80
@makezur
Kirill Mazur
2 years
Survival of the fittest!
@marwan_ptr
Marwan Taher
2 years
Excited to announce Fit-NGP which will be presented in #ICRA2024! Fit-NGP accurately estimates 6-DoF object poses (~ 1.6mm) leveraging Instant-NGP's density field. With @alzugarayign & @AjdDavison. Project page: https://t.co/RMMz4hKthj Video: https://t.co/5W9LRmJtxq (1/3)
0
0
8
@BaeGwangbin
Gwangbin Bae
2 years
Should the complexity of 3D reconstruction tasks be dependent on # of pixels or # of โ€œthingsโ€ in the image? We use surface normal and segmentation priors to split the image into 2.5D segments and show how this representation can help tackle a wide range of 3D reconstruction tasks
@makezur
Kirill Mazur
2 years
SuperPrimitives enable 3D scene reconstruction at a level of local image regions rather than pixels. With SuperPrimitves, 3D reconstruction is done simply, thanks to powerful off-the-shelf 2D priors. project page: https://t.co/l360soWiAz video: https://t.co/H3GJ81qxx2
1
7
31
@makezur
Kirill Mazur
2 years
SuperPrimitives are constructed by splitting an image into a set of regions with a segmentation network. Think of it as a modern analogue of *superpixels*! To get its 3D shape, each segment is enhanced with surface normals, predicted by another network.
0
0
6
@makezur
Kirill Mazur
2 years
joint work with @BaeGwangbin and @AjdDavison
1
0
4
@makezur
Kirill Mazur
3 years
Our work on leveraging general pre-trained feature extractors for open-set real-time scene understanding has been accepted for #ICRA2023! Joint work with @SucarEdgar @AjdDavison Check out project page:
@AjdDavison
Andrew Davison
3 years
New: Feature-realistic neural fusion for real-time, open set scene understanding. Our neural field renders to feature space, enabling real-time grouping and segmentation of similar objects or parts from ultra-sparse, online interaction. Dyson Robotics Lab. https://t.co/vUn96ubJJe
0
6
44
@makezur
Kirill Mazur
4 years
Can you get SoTA results on 4 various point cloud processing tasks with a single block? The answer is yes! ๐ŸŒProject page: https://t.co/9uHrNARe60 ๐Ÿ“ฝ๏ธVideo: https://t.co/EPSzcyw1po ๐ŸงPaper: https://t.co/uTddO6cntc Join us on Session 8b at #ICCV21 on Oct 15th
2
1
23
@ilyazakharkin
Ilya Zakharkin
4 years
Excited to share our work on Point-Based Modeling of Human Clothing accepted to ICCV'21! ๐Ÿฅณ ๐ŸŒ Project page: https://t.co/IL7M3TFiwa ๐ŸŽฌ Video: https://t.co/b8ZTkMiV0v ๐Ÿ“ Paper: https://t.co/gfQ0kg3cg8 Join us on Session 11 at https://t.co/P1jSQSjbi6 on Oct 13th and 15th
4
9
67