Laura Leal-Taixe
@lealtaixe
Followers
11K
Following
2K
Media
69
Statuses
515
Senior Research Manager at @NVIDIA. Prev Professor at @TU_Muenchen. Computer Vision mostly. Views are my own.
Joined June 2016
RT @GuillemBraso: Excited to share what we've been working on! SeNaTra introduces a backbone where segmentation emerges natively by replac….
0
15
0
RT @AljosaOsep: Turns out that if you learn to downsample (rather than using uniform grid pooling) in Vision Transformers, you no longer ne….
0
9
0
0
0
8
Coolest results on zero-shot, text-supervised semantic segmentation as well as a new kid in town for supervised semantic segmentation: the native segmentation network, the first encoder-only model capable of competing with Mask2former and the other big ones.
1
0
9
The coolest thing is that segmentation emerges even from Imagenet pre-training!
1
1
11
The secret sauce is a learned spatial grouping layer, which computes soft token assignments. The cool thing is that this enables principled feature upsampling, from masks to pixels in the encoder itself! No more heavy decoders needed!
1
1
14
Our work shows that segmentation can be inherently encoded in a model’s internal representations rather than delegated to specialized decoder modules, opening new directions in segmentation-centric backbone architectures.
1
2
9
The time for new architectures is over? Not quite! SeNaTra, a native segmentation backbone, is waiting, let's see how it works 🧵
3
39
205
RT @zhenjun_zhao: A Guide to Structureless Visual Localization. Vojtech Panek, Qunjie Zhou, Yaqing Ding, Sérgio Agostinho, @ZKukelova, @Sat….
0
16
0
RT @aycatakmaz: Thanks @_akhaliq for sharing! During my internship at @NVIDIAAI, we explored zero-shot panoptic completion of Lidar scans —….
0
12
0
RT @NVIDIAAIDev: Spatial AI is increasingly important, and the newest papers from #NVIDIAResearch, 3DGRT and 3DGUT, represent significant a….
0
51
0
RT @zhenjun_zhao: MATCHA:Towards Matching Anything. @FeiXue94, @s_elflein, @lealtaixe, @QunjieZhou. tl;dr: diffusion model->semantic+geomet….
0
29
0
RT @JennySeidensch1: If you want to try out our 3DV paper #DynOMo for dynamic, online, monocular reconstruction-based point tracking, you c….
0
4
0
To appear at 3DV! Congrats to the team, especially @JennySeidensch1 !.
You wondered how point tracks generated from dynamic, online, monocular reconstruction look in action? Enjoy the sneak peak of #DynOMo on TAPVID-Davis, PanopticSports and the iPhone dataset! More visuals soon 🚀@lealtaixe @QunjieZhou @BDuisterhof @RamananDeva
0
2
59
My group at NVIDIA is looking for interns for next year! If you are excited about research in dynamic scene understanding, perception & reconstruction, camera pose estimation, or even climate modelling, apply using the link below or contact me directly!
10
82
438
The only problem of choosing such a wonderful group of people is that know I owe a lot of favours. So if you ever see me as PC of a big conference, you can assume one of these favours was paid 😆😆.
4
0
18
They put their heart and soul into preparing a technical program for 7k people. Most do not know how draining this is, and again offers criticisms like they are giving out candy. I can only advice to thank the PCs of a conference, always, thank them. They deserve this and more.
1
0
19
For future organizers: choose your team wisely. Our best decision was to choose a wonderful group of PCs! Sharing this journey with them was the best part of ECCV. @gulvarol @orussakovsky @eliricci_ @stefanroth @SattlerTorsten and Ales.
1
0
22