roboticist, postdoc @ ETH Zürich
It takes a while to make fancy #NeRF animations, so I am very happy we can now share our upcoming #CVPR paper with video and code release:
A big debate in #ContinualLearning is how to scale to many experiences. This work shows how well NeRF-based compression can scale to store robotic experiences over many consecutive deployments, much better than storing checkpoints of your model.
website: https://ethz-asl.github.io/ucsa_neural_rendering/
#Robotics #SceneUnderstanding
A little thing that I enjoy every time I walk into the lab: We got yellow filament and can now print all our robot addons in matching colors 😃🦾
@ducha_aiki My guess would be that most robotics systems are still on opencv 3 😃
eg there is no opencv4-catkin
Statement from the listed authors of Stochastic Parrots on the “AI pause” letter
https://www.dair-institute.org/blog/letter-statement-March2023
"Regulatory efforts should focus on transparency, accountability and preventing exploitative labor practices."
With @timnitGebru @meg and Angelina McMillan-Major
This argues that - because of the "perils of a public good in private hands" - not just discussion should move from twitter to mastodon, scholarly institutions should now also create instances in the fediverse that make publicly available: papers, data & code.
https://www.nature.com/articles/d41586-023-00486-3
@mundt_martin A good keyboard is such a joy.
Excited for my first day of „official“ teaching today: A seminar that will, over the course of the semester, cover milestone papers of the ~10 deep-learning years in computer vision, as well as the issues of uncertainty, biases, and calibration.
academics: digital infrastructure is not my job, my job is to do science
also academics: (spend 80% of time compensating for infrastructural deficits re-solving the same problems privately and struggling with shitty homebrew/cloud experimental, data, and analytical systems)
Just obtained a PDF of the article that (as far as I know) for the first time in the scientific literature uses the phrase "Publish or perish". It is from 1927 - but if you'd publish it again in 2027, I think no one would notice 😀
Case, C. M. (1927). Scholarship in sociology. Sociology and Social Research, 12, 323–340.
NICER-SLAM: Neural Implicit Scene Encoding for RGB SLAM
Zihan Zhu, Songyou Peng, Viktor Larsson, Zhaopeng Cui, Martin R. Oswald, Andreas Geiger, Marc Pollefeys
tl;dr: from RGBD to RGB input: add monodepth, mononormals and optical flow + hierarchical grid
vMAP: Vectorised Object Mapping for Neural Field SLAM
Xin Kong, Shikun Liu, Marwan Taher, Andrew J. Davison
tl;dr: NERF-SLAM + instance segmentation -> per-object MLP decomposable representation
I don't see a whole lot of aerial robots, mobile robots, or even autonomous cars on here. There anyone I should be following?
@micahcorah Lately I‘m tying out @ivory as a phone client and @buffer for scheduling posts. But no conclusions yet.
The IEEE RAS Women in Engineering Committee (WiE) wants to support the community and people interested in robotics and automation research by introducing new travel support that encourages Inclusion, Diversity, Equity, and Accessibility (IDEA).
Learn more: https://www.ieee-ras.org/about-ras/latest-news/ras-wie-announcing-idea-a-new-form-of-travel-support
Continuing our paper perception polls.
What is your personal estimate of CONDITIONAL probability (NOT POSTERIOR) that paper is correct, given it is accepted to #CVPR?
Part of a great conference is a well organized submission process. That‘s why I am getting more and more disappointed with RSS. They announced a demo track, but wanted to publish detailed submission info only a month before the deadline. This Jan 4 announcement was missed and officially pushed to Jan 8, which they also missed. Today we have 2 weeks before the submission deadline and still no info about the demo track. And no response to emails to the chair.
Details for the Continual Learning #CLVision workshop at #CVPR2023 are announced!
🗨️ exciting line-up of speakers
📜 call for papers (deadline now 13 March), with archival and non-archival track
🔥 a challenge (coming soon)
📅 date of workshop: 18 June
https://sites.google.com/view/clvision2023
As it is tradition in many robotics labs, my old lab just published their Christmas video: https://youtu.be/x8wJ1hO_Pcs
Happy Holidays!
RT @osanseviero@twitter.com
OpenAI released Point-E, a text-to-3D (point clouds) demo 🔥
You can check out an open-source demo for it at https://hf.co/spaces/osanseviero/point-e🤯Enjoy! The demo uses the lower-quality but much faster version of the model.
🐦🔗: https://twitter.com/osanseviero/status/1605227395581100032
Yesterday, our students had their big demo day showing off their concepts on how to link robots to mixed reality devices. Lots of cool ideas!
