- 17
- 94 137
SU Lab UC San Diego
Приєднався 3 кві 2020
SU lab at UC San Diego
Відео
Learning Generic and Generalizable Object Manipulation Policies
Переглядів 1,1 тис.2 роки тому
To build robots with general task-solving abilities as humans, as a pre-requisite, robots must possess a diverse set of object manipulation skills (generic), and these skills must apply to objects and configurations that are even unseen (generalizable).To foster reproducible, low-cost, and fast-cycle research, Su Lab has been pushing the development of open-source task suites, ManiSkill, as a c...
SAPIEN Manipulation Skill Challenge 2022
Переглядів 8752 роки тому
SAPIEN Manipulation Skill Challenge 2022 Website: sapien.ucsd.edu/challenges/maniskill/2022/ Code: github.com/haosulab/ManiSkill2 Follow us on twitter: HaoSuLabUCSD
Tutorial: Radiance Field Models for Photorealistic Rendering
Переглядів 1,2 тис.2 роки тому
Recently, the neural radiance field (NeRF) and other radiance field (RF) models have achieved great success for generating realistic novel views from a set of 2D images. This tutorial covers the basic concepts related to NeRF. We will also talk about recent advances on RF models as well as delve deeper into implementation techniques. Outline: Recent Advances on Neural Rendering Background: Radi...
Building and Working in Environments for Embodied AI
Переглядів 2,1 тис.2 роки тому
This tutorial provides the basic knowledge to guide researchers, especially those from vision and machine learning backgrounds, to use and build virtual environments for embodied AI research. Tutorial website: ai-workshops.github.io/building-and-working-in-environments-for-embodied-ai-cvpr-2022/ 0:00 Overview of Embodied AI 27:58 The Basic Frameworks and techniques for Embodied AI 1:16:53 Desig...
![[SGP-2022] Deep Learning on Point Clouds](http://i.ytimg.com/vi/gm_oW0bdzHs/mqdefault.jpg)
[SGP-2022] Deep Learning on Point Clouds
Переглядів 30 тис.2 роки тому
Point cloud is an important type of geometric data structure. They are simple and unified structures that avoid the combinatorial irregularities and complexities of meshes. These properties make point clouds widely used for 3D reconstruction or visual understanding applications, such as AR, autonomous driving, and robotics. This course will teach how we apply deep learning methods to point clou...
ICLR 2022 Workshop on Generalizable Policy Learning in the Physical World
Переглядів 6182 роки тому
ai-workshops.github.io/generalizable-policy-learning-in-the-physical-world/
3D Learning for Manipulation: Simulation, Benchmark, and Learning
Переглядів 7083 роки тому
3D Learning for Manipulation: Simulation, Benchmark, and Learning
SAPIEN Open-Source Manipulation Skill Challenge 2021 (New Video)
Переглядів 1,5 тис.3 роки тому
New video for the SAPIEN Open-Source Manipulation Skill Challenge 2021! Website: sapien.ucsd.edu/challenges/ma... Paper: arxiv.org/abs/2107.14483 Github: github.com/haosulab/ManiSkill | github.com/haosulab/ManiSkill-Learn
MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo (ICCV 2021)
Переглядів 5573 роки тому
MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo (ICCV 2021) Authors: Anpei Chen*, Zexiang Xu*, Fuqiang Zhao, Xiaoshuai Zhang, Fanbo Xiang, Jingyi Yu, Hao Su Project website: apchenstu.github.io/mvsnerf/ Paper: arxiv.org/abs/2103.15595 Code: github.com/apchenstu/mvsnerf
SAPIEN Open-Source Manipulation Skill Challenge 2021 (obsolete)
Переглядів 2803 роки тому
(obsolete) Checkout our new video here: ua-cam.com/video/Lu8DJlnvYkA/v-deo.html
Compositional Generalizability in Geometry, Physics, and Policy Learning
Переглядів 6574 роки тому
It is well known that deep neural networks are universal function approximators and have good generalizability when the training and test datasets are sampled from the same distribution. Most deep learning-based applications and theories in the past decade are based upon this setup. While the view of learning function approximators has been rewarding to the community, we are seeing more and mor...
When uploading videos publicly, it's important to ensure good quality. Starting from this point [link to the video](ua-cam.com/video/vfL6uJYFrp4/v-deo.html), it became difficult to understand, and I felt like I wasted my time trying to follow your explanation on 3D segmentation.
This is such a good resource on point cloud. Thank you for uploading
Can you give any instruction on how to apply this to a different robot arm please?
6:45 which paper used shape prior for MVS?
This is great
Thank you
Is there a paper based on this video?
What camera do you recommend for taking photos and creating a 3D point cloud?
Zivid
👍*100
Maybe reprocess the audio or use an automated reader ? Try a condenser microphone. . Quality professional information needs clarity and a professional presentation !
Bunny box. Very important.🤣
Hello sir I am research scholar registered for phD.i am looking for stepwise latest algorithms and architecture for research in poin cloud. will you help please?
Hello Su, What does normal information means that the point cloud comes with?
normals are commonly used with point clouds for rendering. its just a x, y, z normal vector that points toward the direction the point is seeing
Thank you for the video, really nice one, although the second person talking about segmentation has to fix his microphone, he was unmanageable to listen to
Thank you for this talk. The density of information is overwhelming and so I'll be coming back to this a few times!
Thank you for the presentation. Why parts of the presentation are in Chinese without subtitles?
are you chinese ?
Thanks for the great talk.
Thanks You
Thank you .this is so much helpful and informative.
Our brain uses metaheursitics to solve vision problems all time
36:40 - Segmentation and Detection
Can you share this ppt to me If possible because I need this kartikgondaliya0@gmail.com
Tank you! This was very helpful for understanding different 3D learning approaches.
please explain pointCNN. Existing literature is very limited and extremely tough to understand it
medium.com/@luis_gonzales/an-in-depth-look-at-pointnet-111d7efdaa1a I feel this blog explain pointCNN very clearly. Hopefully help you
@@yunshengluo1196 Thank you for the link Luo, but it seems that link is for pointnet. I will highly appreciate if you can help me with any link that has simple explanation to pointcnn