Zobrazeno 1 - 10
of 27
pro vyhledávání: '"Gkanatsios, Nikolaos A."'
Diffusion policies are conditional diffusion models that learn robot action distributions conditioned on the robot and environment state. They have recently shown to outperform both deterministic and alternative action distribution learning formulati
Externí odkaz:
http://arxiv.org/abs/2402.10885
Autor:
Yang, Brian, Su, Huangyuan, Gkanatsios, Nikolaos, Ke, Tsung-Wei, Jain, Ayush, Schneider, Jeff, Fragkiadaki, Katerina
Diffusion models excel at modeling complex and multimodal trajectory distributions for decision-making and control. Reward-gradient guided denoising has been recently proposed to generate trajectories that maximize both a differentiable reward functi
Externí odkaz:
http://arxiv.org/abs/2402.06559
Autor:
Jain, Ayush, Katara, Pushkal, Gkanatsios, Nikolaos, Harley, Adam W., Sarch, Gabriel, Aggarwal, Kriti, Chaudhary, Vishrav, Fragkiadaki, Katerina
State-of-the-art models on contemporary 3D segmentation benchmarks like ScanNet consume and label dataset-provided 3D point clouds, obtained through post processing of sensed multiview RGB-D images. They are typically trained in-domain, forego large-
Externí odkaz:
http://arxiv.org/abs/2401.02416
3D perceptual representations are well suited for robot manipulation as they easily encode occlusions and simplify spatial reasoning. Many manipulation tasks require high spatial precision in end-effector pose prediction, which typically demands high
Externí odkaz:
http://arxiv.org/abs/2306.17817
Autor:
Gkanatsios, Nikolaos, Jain, Ayush, Xian, Zhou, Zhang, Yunchu, Atkeson, Christopher, Fragkiadaki, Katerina
Language is compositional; an instruction can express multiple relation constraints to hold among objects in a scene that a robot is tasked to rearrange. Our focus in this work is an instructable scene-rearranging framework that generalizes to longer
Externí odkaz:
http://arxiv.org/abs/2304.14391
Autor:
Gkanatsios, Nikolaos, Singh, Mayank, Fang, Zhaoyuan, Tulsiani, Shubham, Fragkiadaki, Katerina
We present Analogical Networks, a model that encodes domain knowledge explicitly, in a collection of structured labelled 3D scenes, in addition to implicitly, as model parameters, and segments 3D object scenes with analogical reasoning: instead of ma
Externí odkaz:
http://arxiv.org/abs/2304.14382
Most models tasked to ground referential utterances in 2D and 3D scenes learn to select the referred object from a pool of object proposals provided by a pre-trained detector. This is limiting because an utterance may refer to visual entities at vari
Externí odkaz:
http://arxiv.org/abs/2112.08879
Inherent morphological characteristics in objects may offer a wide range of plausible grasping orientations that obfuscates the visual learning of robotic grasping. Existing grasp generation approaches are cursed to construct discontinuous grasp maps
Externí odkaz:
http://arxiv.org/abs/2006.05123
Autor:
Gkanatsios, Nikolaos, Pitsikalis, Vassilis, Koutras, Petros, Zlatintsi, Athanasia, Maragos, Petros
Detecting visual relationships, i.e. triplets, is a challenging Scene Understanding task approached in the past via linguistic priors or spatial information in a single feature branch. We introduce a new deeply supervised
Externí odkaz:
http://arxiv.org/abs/1902.05829
3D perceptual representations are well suited for robot manipulation as they easily encode occlusions and simplify spatial reasoning. Many manipulation tasks require high spatial precision in end-effector pose prediction, typically demanding high-res
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2323d4ba0cbafc63d37ad85629549612
http://arxiv.org/abs/2306.17817
http://arxiv.org/abs/2306.17817