Zobrazeno 1 - 10
of 36
pro vyhledávání: '"Schmeckpeper, Karl"'
Autor:
Biza, Ondrej, Weng, Thomas, Sun, Lingfeng, Schmeckpeper, Karl, Kelestemur, Tarik, Ma, Yecheng Jason, Platt, Robert, van de Meent, Jan-Willem, Wong, Lawson L. S.
Reinforcement Learning (RL) has the potential to enable robots to learn from their own actions in the real world. Unfortunately, RL can be prohibitively expensive, in terms of on-robot runtime, due to inefficient exploration when learning from a spar
Externí odkaz:
http://arxiv.org/abs/2410.19989
Autor:
Shang, Jinghuan, Schmeckpeper, Karl, May, Brandon B., Minniti, Maria Vittoria, Kelestemur, Tarik, Watkins, David, Herlant, Laura
Vision-based robot policy learning, which maps visual inputs to actions, necessitates a holistic understanding of diverse visual tasks beyond single-task needs like classification or segmentation. Inspired by this, we introduce Theia, a vision founda
Externí odkaz:
http://arxiv.org/abs/2407.20179
Autor:
Huang, Haojie, Schmeckpeper, Karl, Wang, Dian, Biza, Ondrej, Qian, Yaoyao, Liu, Haotian, Jia, Mingxi, Platt, Robert, Walters, Robin
Humans can imagine goal states during planning and perform actions to match those goals. In this work, we propose Imagination Policy, a novel multi-task key-frame policy network for solving high-precision pick and place tasks. Instead of learning act
Externí odkaz:
http://arxiv.org/abs/2406.11740
Despite outstanding semantic scene segmentation in closed-worlds, deep neural networks segment novel instances poorly, which is required for autonomous agents acting in an open world. To improve out-of-distribution (OOD) detection for segmentation, w
Externí odkaz:
http://arxiv.org/abs/2311.07578
We introduce Equivariant Neural Field Expectation Maximization (EFEM), a simple, effective, and robust geometric algorithm that can segment objects in 3D scenes without annotations or training on scenes. We achieve such unsupervised segmentation by e
Externí odkaz:
http://arxiv.org/abs/2303.15440
Autor:
Schmeckpeper, Karl, Osteen, Philip R., Wang, Yufu, Pavlakos, Georgios, Chaney, Kenneth, Jordan, Wyatt, Zhou, Xiaowei, Derpanis, Konstantinos G., Daniilidis, Kostas
Publikováno v:
Field Robotics, 2, 147-171, 2022
This paper presents an approach to estimating the continuous 6-DoF pose of an object from a single RGB image. The approach combines semantic keypoints predicted by a convolutional network (convnet) with a deformable shape model. Unlike prior investig
Externí odkaz:
http://arxiv.org/abs/2204.05864
Autor:
Georgakis, Georgios, Schmeckpeper, Karl, Wanchoo, Karan, Dan, Soham, Miltsakaki, Eleni, Roth, Dan, Daniilidis, Kostas
We consider the problem of Vision-and-Language Navigation (VLN). The majority of current methods for VLN are trained end-to-end using either unstructured memory such as LSTM, or using cross-modal attention over the egocentric observations of the agen
Externí odkaz:
http://arxiv.org/abs/2203.05137
Autor:
Georgakis, Georgios, Bucher, Bernadette, Arapin, Anton, Schmeckpeper, Karl, Matni, Nikolai, Daniilidis, Kostas
We consider the problems of exploration and point-goal navigation in previously unseen environments, where the spatial complexity of indoor scenes and partial observability constitute these tasks challenging. We argue that learning occupancy priors o
Externí odkaz:
http://arxiv.org/abs/2202.11907
Autor:
Ebert, Frederik, Yang, Yanlai, Schmeckpeper, Karl, Bucher, Bernadette, Georgakis, Georgios, Daniilidis, Kostas, Finn, Chelsea, Levine, Sergey
Robot learning holds the promise of learning policies that generalize broadly. However, such generalization requires sufficiently diverse datasets of the task of interest, which can be prohibitively expensive to collect. In other fields, such as comp
Externí odkaz:
http://arxiv.org/abs/2109.13396
Autor:
Georgakis, Georgios, Bucher, Bernadette, Schmeckpeper, Karl, Singh, Siddharth, Daniilidis, Kostas
We consider the problem of object goal navigation in unseen environments. Solving this problem requires learning of contextual semantic priors, a challenging endeavour given the spatial and semantic variability of indoor environments. Current methods
Externí odkaz:
http://arxiv.org/abs/2106.15648