Zobrazeno 1 - 10
of 22
pro vyhledávání: '"Gothoskar, Nishad"'
Autor:
Curtis, Aidan, Matheos, George, Gothoskar, Nishad, Mansinghka, Vikash, Tenenbaum, Joshua, Lozano-Pérez, Tomás, Kaelbling, Leslie Pack
Integrated task and motion planning (TAMP) has proven to be a valuable approach to generalizable long-horizon robotic manipulation and navigation problems. However, the typical TAMP problem formulation assumes full observability and deterministic act
Externí odkaz:
http://arxiv.org/abs/2403.10454
Autor:
Gothoskar, Nishad, Ghavami, Matin, Li, Eric, Curtis, Aidan, Noseworthy, Michael, Chung, Karen, Patton, Brian, Freeman, William T., Tenenbaum, Joshua B., Klukas, Mirko, Mansinghka, Vikash K.
Robots cannot yet match humans' ability to rapidly learn the shapes of novel 3D objects and recognize them robustly despite clutter and occlusion. We present Bayes3D, an uncertainty-aware perception system for structured 3D scenes, that reports accur
Externí odkaz:
http://arxiv.org/abs/2312.08715
Autor:
Gothoskar, Nishad
A central challenge in 3D scene perception via inverse graphics is robustly modeling the gap between 3D graphics and real-world data. We propose a novel 3D Neural Embedding Likelihood (3DNEL) over RGB-D images to address this gap. 3DNEL uses neural e
Externí odkaz:
https://hdl.handle.net/1721.1/150082
Autor:
Zhou, Guangyao, Gothoskar, Nishad, Wang, Lirui, Tenenbaum, Joshua B., Gutfreund, Dan, Lázaro-Gredilla, Miguel, George, Dileep, Mansinghka, Vikash K.
The ability to perceive and understand 3D scenes is crucial for many applications in computer vision and robotics. Inverse graphics is an appealing approach to 3D scene understanding that aims to infer the 3D scene structure from 2D images. In this p
Externí odkaz:
http://arxiv.org/abs/2302.03744
Autor:
Zhi-Xuan, Tan, Gothoskar, Nishad, Pollok, Falk, Gutfreund, Dan, Tenenbaum, Joshua B., Mansinghka, Vikash K.
To facilitate the development of new models to bridge the gap between machine and human social intelligence, the recently proposed Baby Intuitions Benchmark (arXiv:2102.11938) provides a suite of tasks designed to evaluate commonsense reasoning about
Externí odkaz:
http://arxiv.org/abs/2208.02914
Autor:
Gothoskar, Nishad, Lázaro-Gredilla, Miguel, Bekiroglu, Yasemin, Agarwal, Abhishek, Tenenbaum, Joshua B., Mansinghka, Vikash K., George, Dileep
Visual servoing enables robotic systems to perform accurate closed-loop control, which is required in many applications. However, existing methods either require precise calibration of the robot kinematic model and cameras or use neural architectures
Externí odkaz:
http://arxiv.org/abs/2202.03697
Autor:
Gothoskar, Nishad, Cusumano-Towner, Marco, Zinberg, Ben, Ghavamizadeh, Matin, Pollok, Falk, Garrett, Austin, Tenenbaum, Joshua B., Gutfreund, Dan, Mansinghka, Vikash K.
We present 3DP3, a framework for inverse graphics that uses inference in a structured generative model of objects, scenes, and images. 3DP3 uses (i) voxel models to represent the 3D shape of objects, (ii) hierarchical scene graphs to decompose scenes
Externí odkaz:
http://arxiv.org/abs/2111.00312
For an intelligent agent to flexibly and efficiently operate in complex environments, they must be able to reason at multiple levels of temporal, spatial, and conceptual abstraction. At the lower levels, the agent must interpret their proprioceptive
Externí odkaz:
http://arxiv.org/abs/2006.06620
Autor:
Lázaro-Gredilla, Miguel, Lehrach, Wolfgang, Gothoskar, Nishad, Zhou, Guangyao, Dedieu, Antoine, George, Dileep
Probabilistic graphical models (PGMs) provide a compact representation of knowledge that can be queried in a flexible way: after learning the parameters of a graphical model once, new probabilistic queries can be answered at test time without retrain
Externí odkaz:
http://arxiv.org/abs/2006.06803
Autor:
Gothoskar, Nishad, Lázaro-Gredilla, Miguel, Agarwal, Abhishek, Bekiroglu, Yasemin, George, Dileep
We introduce a novel formulation for incorporating visual feedback in controlling robots. We define a generative model from actions to image observations of features on the end-effector. Inference in the model allows us to infer the robot state corre
Externí odkaz:
http://arxiv.org/abs/2003.04474