2024.10.15 - #11 - CLIP-Clique, FoundPose, Tesla CyberCab/Robovan, MEVIUS, WildFusion, 𝛼LiDAR

[changh95]

Interesting papers

CLIP-Clique: Graph-based Correspondence Matching Augmented by Vision Language Models for Object-based Global Localization

FoundPose: Unseen Object Pose Estimation with Foundation Features

• https://arxiv.org/pdf/2311.18809

Industry news

Tesla CyberCab & Robovan

• 2027년 전에 출시 예정
• $30000
• "fully autonomous unsupervised"

Hiring

• Northwestern University, Tenure-Track Assistant Professor in Embodied Artificial Intelligence 링크

[U-AMC]

Interesting papers

MEVIUS

https://github.com/haraduka/mevius

WildFusion: Multimodal Implicit 3D Reconstructions in the Wild

https://github.com/generalroboticslab/WildFusion

𝛼LiDAR: An Adaptive High-Resolution Panoramic LiDAR System

https://github.com/HViktorTsoi/alpha_lidar

Learning Smooth Humanoid Locomotion through Lipschitz-Constrained Policies

https://lipschitz-constrained-policy.github.io/