Edgar Sucar
@SucarEdgar
Followers
778
Following
8K
Media
11
Statuses
112
Postdoc @Oxford_VGG | PhD Dyson Robotics Lab at Imperial College
Oxford, UK
Joined April 2017
The only bitter lesson is that LLMs have succeeded beyond any expert expectations. Underpinning LLMs is the idea of scaling, which is too often misunderstood as more parameters. Scaling is about using massive compute effectively to maximise the throughput of data ingestion into
@GaryMarcus @ylecun @demishassabis You were never alone, Gary, though you were the first to bite the bullet, to fight the good fight, and to make the argument well, again and again, for the limitations of LLMs. I salute you for this good service!
39
76
698
Good essay on the analogy of the stone soup tale to AI misconception. More emphasis is placed on individual AI models and teams/algorithms who made them, rather than on the collective effort to generate big data, the most important ingredient of the soup. https://t.co/Zebm1Qj1ft
simons.berkeley.edu
For some time, I’ve argued that a common conception of AI is misguided. This is the idea that AI systems like large language and vision models are individual intelligent agents, analogous to human...
0
1
7
SLAM bitter lesson: methods that do less “test time optimisation” will eventually trump over the methods that do more .
4
0
19
Among the keynote speakers, we had two great keynotes delivered by young researchers: Dr. Saiph Savage @saiphcita and Dr. Edgar Sucar @SucarEdgar , who besides being excellent researchers, are proudly Mexican!
1
2
6
Delighted to share the first project of my PhD, "Transient Neural Radiance Fields for Lidar View Synthesis and 3D Reconstruction". We show unprecedented capabilities of synthesizing novel lidar scans from as few as 2 input views! 🖥️ https://t.co/rpOB2ar1F3
6
41
180
Congratulations to Edgar who passed his PhD viva today, and thanks to examiners @tolga_birdal and José María Montiel! A reminder of Edgar's iMAP, a landmark as the first real-time neural field SLAM system from #ICCV2021.
Excited to share iMAP, first real-time SLAM system to use an implicit scene network as map representation. Work with: @liu_shikun, @joeaortiz, @AjdDavison Project page: https://t.co/Tagk4jFN2M Paper: https://t.co/OQA1QdLY4Q
3
3
84
vMAP: Vectorised Object Mapping for Neural Field SLAM, new at #CVPR2023! Each object is represented by a separate MLP, optimised in parallel via vectorised training. Full info: https://t.co/I6dVakYXik Dyson Robotics Lab, Imperial College, @liu_shikun, Marwan Taher, @AjdDavison.
4
27
126
Excited to announce "Learning a Depth Covariance Function" with @AjdDavison. A flexible framework for a variety of geometric vision tasks, such as dense monocular visual odometry shown below. Dyson Robotics Lab, Imperial College Project page: https://t.co/akHcNZztZX
#CVPR2023
6
34
156
What is ChatGPT doing ... and why does it work? From the lore of neural nets to what Aristotle didn't get to ... here's my version of the story: https://t.co/J1YVJvumEp
73
1K
5K
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:
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
If you're at CORL this week in NZ come and meet @iainhaughton and @Ed__Johns and see Iain's presentation in the oral session on Saturday. Dense, fully automatic segmentation of scene properties like softness via real-time neural field training; no priors! It's iLabel for robots.
#CORL2022 oral: mapping non-visual properties (material, softness, force) from very few point sensor tests. iLabel-like neural field produces dense maps + guides actions. Live, autonomous, no priors! @iainhaughton et al, @Dyson @ICComputing. Paper/video: https://t.co/hCR0ko8U16
0
3
35
#CORL2022 oral: mapping non-visual properties (material, softness, force) from very few point sensor tests. iLabel-like neural field produces dense maps + guides actions. Live, autonomous, no priors! @iainhaughton et al, @Dyson @ICComputing. Paper/video: https://t.co/hCR0ko8U16
2
15
112
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
2
39
281
"A Year" AI animation artwork made in colab using my custom stable 3D animation algorithm on top of #stablediffusion model. In the thread I share some details about the algo and when i plan to release it, and talk about the joy and future of AI filmmaking 🎶 DakhaBrakha - Vesna
136
986
6K
cool work by Dominic Maggio and collaborators on real-time robot localization using Neural Radiance Fields (NeRF) with Monte Carlo Localization: https://t.co/jG9Vp936g6 (link to full paper in the video description) #mitSparkLab #NeRF #neuralFields #robotics #localization
4
18
129
“..the actual contents of minds are tremendously, irredeemably complex; we should stop trying to find simple ways to think about the contents of minds, such as simple ways to think about space, objects, multiple agents, or symmetries. ”
0
0
2