• Publications
  • Influence
RoboTHOR: An Open Simulation-to-Real Embodied AI Platform
TLDR
RoboTHOR offers a framework of simulated environments paired with physical counterparts to systematically explore and overcome the challenges of simulation-to-real transfer, and a platform where researchers across the globe can remotely test their embodied models in the physical world.
Naive Bayes Super-Resolution Forest
TLDR
A bimodal tree for clustering, which successfully exploits the antipodal invariance of the coarse-to-high-res mapping of natural image patches and provides scalability to finer partitions of the underlying coarse patch space, is presented.
Non-Parametric Blur Map Regression for Depth of Field Extension
TLDR
This paper presents a blind deblurring pipeline able to restore real camera systems by slightly extending their DOF and recovering sharpness in regions slightly out of focus by relying first on the estimation of the spatially varying defocus blur.
PSyCo: Manifold Span Reduction for Super Resolution
TLDR
A novel regression-based SR algorithm that benefits from an extended knowledge of the structure of both manifolds, and proposes a transform that collapses the 16 variations induced from the dihedral group of transforms and antipodality into a single primitive.
Image-based multi-view scene analysis using 'conexels'
TLDR
A 3D geometry is proposed for multi-view scene analysis providing a better balance in terms of the number of pixels used to analyse each elementary volumetric unit, which is non-regular in 3D space, but becomes regular once projected onto camera images, adapting the sampling to the images.
AllenAct: A Framework for Embodied AI Research
TLDR
AllenAct is introduced, a modular and flexible learning framework designed with a focus on the unique requirements of Embodied AI research that provides first-class support for a growing collection of embodied environments, tasks and algorithms.
Antipodally Invariant Metrics for Fast Regression-Based Super-Resolution
TLDR
This paper presents a very fast regression-based algorithm, which builds on the densely populated anchored neighborhoods and sublinear search structures, and proposes a simple yet effective antipodally invariant transform that can be easily included in the Euclidean distance calculation.
Learning About Objects by Learning to Interact with Them
TLDR
This work presents a computational framework to discover objects and learn their physical properties along this paradigm of Learning from Interaction, and reveals that this agent learns efficiently and effectively; not just for objects it has interacted with before, but also for novel instances from seen categories as well as novel object categories.
Fast Super-Resolution via Dense Local Training and Inverse Regressor Search
TLDR
This paper proposes a novel inverse-search approach for regression-based Super-Resolution by applying spherical hashing to both image and regressors, which reduces the inverse search into computing a trained function.
Half hypersphere confinement for piecewise linear regression
TLDR
The usage of antipodally invariant metrics are proposed and introduced and the Half Hypersphere Confinement (HHC), a fast alternative to Multidimensional Scaling (MDS) that allows to map antipodal invariant distances in the Euclidean space with very little approximation error is introduced.
...
1
2
3
4
5
...