Home /
Research
Showing 247 - 252 / 897
Rethinking Text Segmentation: A Novel Dataset and A Text-Specific Refinement Approach
XingqianXu, ZhifeiZhang, ZhaowenWang....
Published date-11/27/2020
SceneText, StyleTransfer, TextStyleTransfer
Text segmentation is a prerequisite in many real-world text-related tasks, e.g., text style transfer, and scene text removal. However, facing the lack of high-quality datasets and dedicated investigations, this critical …
Efficient Scene Compression for Visual-based Localization
MarcelaMera-Trujillo, BenjaminSmith, VictorFragoso....
Published date-11/27/2020
3DReconstruction
Estimating the pose of a camera with respect to a 3D reconstruction or scene representation is a crucial step for many mixed reality and robotics applications. Given the vast amount …
Domain Adaptative Causality Encoder
FarhadMoghimifar, GholamrezaHaffari, MahsaBaktashmotlagh....
Published date-11/27/2020
Current approaches which are mainly based on the extraction of low-level relations among individual events are limited by the shortage of publicly available labelled data. Therefore, the resulting models perform …
Generalized Pose-and-Scale Estimation using 4-Point Congruence Constraints
VictorFragoso, SudiptaSinha....
Published date-11/27/2020
We present gP4Pc, a new method for computing the absolute pose of a generalized camera with unknown internal scale from four corresponding 3D point-and-ray pairs. Unlike most pose-and-scale methods, gP4Pc …
Simulation-efficient marginal posterior estimation with swyft: stop wasting your precious time
BenjaminKurtMiller, AlexCole, GillesLouppe....
Published date-11/27/2020
BayesianInference
We present algorithms (a) for nested neural likelihood-to-evidence ratio estimation, and (b) for simulation reuse via an inhomogeneous Poisson point process cache of parameters and corresponding simulations. Together, these algorithms …
Point and Ask: Incorporating Pointing into Visual Question Answering
ArjunMani, WillHinthorn, NoblineYoo....
Published date-11/27/2020
QuestionAnswering, VisualQuestionAnswering
Visual Question Answering (VQA) has become one of the key benchmarks of visual recognition progress. Multiple VQA extensions have been explored to better simulate real-world settings: different question formulations, changing …