@sim2realAIorg

We are indexing the progress in simulations to real world transfer in perception and control. https://xitter.cf/t.co/nKLQmmKM46 Managed by @ankurhandos

 London, England
sim2realai.github.io
Joined July 2018
  • Tweets511
  • Following0
  • Followers3.6K
370Media
We introduce a new method to learn simulators from depth and RGB videos. The "URDF" of an articulated rigid-body mechanism is reconstructed, and the parameters of the simulator inferred through Bayesian inference. Website: eric-heiden.github.io/video2… Paper: arxiv.org/abs/2203.10488
4
47
6
300
We have released a super fast implementation of AMP using @nvidia's Isaac Gym. Models that used to take days to train can now be trained in minutes! github.com/NVIDIA-Omniverse/… There's also a retargeting script for loading new motion clips. Thanks to @viktor_m81, @gavrielstate.
2
72
4
476
Super excited to share my internship work at @ToyotaResearch on category-level 3D object understanding and single-shot real2sim asset creation, accepted at #ICRA2022! zubair-irshad.github.io/proj… @GTrobotics @ICatGT @mlatgt @ieee_ras_icra ⬇️(1/6)
2
42
7
217
We at @StandardAI are thrilled to introduce StandardSim, a Photorealistic Synthetic #dataset for #AI research in Retail Environments! 25K+ high-resolution multi-view renders with pixel-perfect labels 🚀 standard-ai.github.io/Standa… arxiv.org/abs/2202.02418
11
58
11
315
We are releasing MOTSynth, the largest dataset for Pedestrian Detection and Tracking. Expect to obtain better real-world performances by training your model on this data only. It's synthetic, but it's for real! lnkd.in/d-NvqiX4
5
80
10
451
Happy to announce that the 3rd Embodied AI Workshop is happening on June 19th at #CVPR2022. Lots of exciting speakers and 13 (!!!) challenges. Check out our website (embodied-ai.org), dive into a challenge, and save the date!
3
34
7
94
Released our work on object extraction: SafePicking #ICRA2022! This robotic system finds and “safely” extracts target objects via known object mapping with an onboard camera. Check out the project page for paper/code: safepicking.wkentaro.com Work with @stepjamUK @AjdDavison (1/n)
1
40
4
116
What I’ve learned about making synthetic data work for training ML models:
14
175
9
1,050
Some simulation platforms that caught my eye this year, 2021 🧶
Ankur Handa
@ankurhandos20 Dec 2020
Some simulation platforms that caught my eye this year 🧵
2
13
1
74
Realistic and diverse simulation is a crucial tool for building AV2.0 - safe, scalable, and adaptable learned driving. If you share our passion for sim and AV, and want to work in a dynamic and diverse team at the forefront of AI, get in touch at [email protected] @wayve_ai.
3
25
2
121
Need Sim2Real baselines? On finishing his PhD with us, Fabio has released version 1.0 of SimuRLacra github.com/famura/SimuRLacra a PyTorch-based RL framework focusing on Sim2Real transfer methods

GitHub - famura/SimuRLacra: reinforcement learning from randomized simulations

reinforcement learning from randomized simulations - famura/SimuRLacra

github.com
15
59
"Simulation Intelligence" is the merger of scientific computing, simulation, and AI to broaden the scientific method for the machine age. arxiv.org/abs/2112.03235
22
176
33
804
How well do today's AI models understand the physical structure and dynamics of visual scenes? Introducing #Physion: a dataset/benchmark featuring objects that roll, slide, fall, fold, collide, connect, contain, and more! preprint: arxiv.org/abs/2106.08261
3
77
6
317
DiffSDFSim: Differentiable Rigid-Body Dynamics With Implicit Shapes arxiv.org/abs/2111.15318 differentiable physics with frictional contacts. infers physical parameters such as friction coefficients, mass, forces or shape parameters from trajectory and depth image observations.
1
21
68
Introduction to Computer Graphics Lecture 25 on "Simulation in Graphics" is online: youtube.com/watch?v=iZ1e_1tn… This last lecture of the course is an overview of various simulation methods used in computer graphics. #graphics #cg
Cem Yuksel@cem_yuksel31 Jan 2021
Introduction to Computer Graphics Recordings of my live online lectures are now on my youtube channel. Here is the playlist: youtube.com/watch?v=vLSphLtK… The first 4 lectures are online. The rest will come soon. #graphics
1
8
48
We developed dynamic environments with highly deformable topologically non-trivial objects; these offer fast runtime, large-scale parallel data generation, easy reinforcement learning integration via OpenAI Gym API. #NeurIPS D&B track openreview.net/forum?id=WcY3…
33
4
128
Sim-to-Real for High-Resolution Optical Tactile Sensing: From Images to Three-Dimensional Contact Force Distributions Read Now: ow.ly/E0mX50GYw2i
5
11
'basketball_in_hoop'; one of many new tasks joining the #RLBench family of 100+ tasks in V1.2. Coming early November! 🤖 RLBench is still the hardest manipulation sim-benchmark to date due to its large-scale focus on vision, sparse rewards, and multi-stage tasks.
16
101
Synthetic Supervision + Self-Supervision = 💕 In their latest blog, our machine learning team shares how TRI is leveraging photorealistic synthetic datasets for dynamic scene understanding, specifically for #autonomousdriving. Check it out: medium.com/toyotaresearch/sc…

Scaling up Synthetic Supervision for Computer Vision

By Jie Li, Vitor Guizilini, Pavel Tokmakov, Sergey Zakharov, and Adrien Gaidon

medium.com
1
25
4
117
Robots Must Be Ephemeralized - in this blog post I recount how 3 yrs ago, I was bearish on robot research in sim. I've changed my mind completely - I now believe that sim becomes *more* indispensable as robots become more general-purpose blog.evjang.com/2021/09/ephe…
7
18
3
110
Load more