Zobrazeno 1 - 10
of 60
pro vyhledávání: '"Ranftl, René"'
We present a visual-inertial depth estimation pipeline that integrates monocular depth estimation and visual-inertial odometry to produce dense depth estimates with metric scale. Our approach performs global scale and shift alignment against sparse m
Externí odkaz:
http://arxiv.org/abs/2303.12134
Semantic segmentation models struggle to generalize in the presence of domain shift. In this paper, we introduce contrastive learning for feature alignment in cross-domain adaptation. We assemble both in-domain contrastive pairs and cross-domain cont
Externí odkaz:
http://arxiv.org/abs/2204.08399
We present LSeg, a novel model for language-driven semantic image segmentation. LSeg uses a text encoder to compute embeddings of descriptive input labels (e.g., "grass" or "building") together with a transformer-based image encoder that computes den
Externí odkaz:
http://arxiv.org/abs/2201.03546
We study the problem of estimating room layouts from a single panorama image. Most former works have two stages: feature extraction and parametric model fitting. Here we propose an end-to-end method that directly predicts parametric layouts from an i
Externí odkaz:
http://arxiv.org/abs/2112.11340
Autor:
Loquercio, Antonio, Kaufmann, Elia, Ranftl, René, Müller, Matthias, Koltun, Vladlen, Scaramuzza, Davide
Publikováno v:
Science Robotics 2021 Vol. 6, Issue 59, abg5810
Quadrotors are agile. Unlike most other machines, they can traverse extremely complex environments at high speeds. To date, only expert human pilots have been able to fully exploit their capabilities. Autonomous operation with on-board sensing and co
Externí odkaz:
http://arxiv.org/abs/2110.05113
Weight sharing promises to make neural architecture search (NAS) tractable even on commodity hardware. Existing methods in this space rely on a diverse set of heuristics to design and train the shared-weight backbone network, a.k.a. the super-net. Si
Externí odkaz:
http://arxiv.org/abs/2110.01154
Weight sharing has become a de facto standard in neural architecture search because it enables the search to be done on commodity hardware. However, recent works have empirically shown a ranking disorder between the performance of stand-alone archite
Externí odkaz:
http://arxiv.org/abs/2104.05309
We introduce dense vision transformers, an architecture that leverages vision transformers in place of convolutional networks as a backbone for dense prediction tasks. We assemble tokens from various stages of the vision transformer into image-like r
Externí odkaz:
http://arxiv.org/abs/2103.13413
Autor:
Kaufmann, Elia, Loquercio, Antonio, Ranftl, René, Müller, Matthias, Koltun, Vladlen, Scaramuzza, Davide
Publikováno v:
Robotics, Science, and Systems (RSS), 2020
Performing acrobatic maneuvers with quadrotors is extremely challenging. Acrobatic flight requires high thrust and extreme angular accelerations that push the platform to its physical limits. Professional drone pilots often measure their level of mas
Externí odkaz:
http://arxiv.org/abs/2006.05768
Many problems in science and engineering can be formulated in terms of geometric patterns in high-dimensional spaces. We present high-dimensional convolutional networks (ConvNets) for pattern recognition problems that arise in the context of geometri
Externí odkaz:
http://arxiv.org/abs/2005.08144