F
Federico Tombari
Researcher at Technische Universität München
Publications - 332
Citations - 17062
Federico Tombari is an academic researcher from Technische Universität München. The author has contributed to research in topics: Computer science & Pose. The author has an hindex of 48, co-authored 278 publications receiving 12522 citations. Previous affiliations of Federico Tombari include École Polytechnique Fédérale de Lausanne & Ludwig Maximilian University of Munich.
Papers
More filters
Posted Content
Go with the Flows: Mixtures of Normalizing Flows for Point Cloud Generation and Reconstruction.
TL;DR: In this paper, a mixture of normalizing flows is used to generate point clouds with improved details compared to single-flow-based models while using fewer parameters and considerably reducing the inference runtime.
Book ChapterDOI
A Radial Search Method for Fast Nearest Neighbor Search on Range Images
TL;DR: When tested against open source implementations of state-of-the-art NNS algorithms, radial search obtains better performance than the other algorithms in terms of speedup, while yielding the same level of accuracy.
Posted Content
Unconditional Scene Graph Generation
TL;DR: In this paper, a deep auto-regressive model called SceneGraphGen is proposed to learn the probability distribution over labeled and directed scene graphs using a hierarchical recurrent architecture, which can generate scene graphs in a sequence of steps, each step generating an object node, followed by a series of relationship edges connecting to the previous nodes.
Journal ArticleDOI
NEWTON: Neural View-Centric Mapping for On-the-Fly Large-Scale SLAM
TL;DR: Newton as mentioned in this paper proposes a view-centric mapping method that dynamically constructs neural fields based on run-time observation, which enables camera pose updates using loop closures and scene boundary updates by representing the scene with multiple neural fields.
Book ChapterDOI
Abstract: Leveraging Web Data for Skin Lesion Classification
TL;DR: The success of deep learning is mainly based on the assumption that for the given application, there is access to a large amount of annotated data, and the representativeness of the training set is still limited by the number of available samples.