the archive

#real-time processing

7 results

cs.CV2019

Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking

Ziyuan Huang, Changhong Fu, Yiming Li +2

The paper introduces an aberrance‑repressed correlation filter (ARCF) that limits changes in response maps to reduce tracking errors, achieving robust, real‑time object tracking fo…

#object tracking#uav vision#correlation filters#real-time processing

cs.CV2019

Quotienting Impertinent Camera Kinematics for 3D Video Stabilization

Thomas W. Mitchel, Christian Wuelker, Jin Seob Kim +2

The paper presents a fast 3D video stabilization framework that uses dense scene flow and a novel camera motion model to directly recover real-world camera poses, handling large mo…

#video stabilization#dense scene flow#camera motion estimation#real-time processing

cs.CV2019

Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidth

Davy Neven, Bert De Brabandere, Marc Proesmans +1

The paper introduces a proposal‑free instance segmentation approach that learns spatial pixel embeddings together with an instance‑specific clustering bandwidth, enabling real‑time…

#instance segmentation#proposal-free methods#spatial embeddings#clustering loss

cs.RO2019

Deep Sensor Fusion for Real-Time Odometry Estimation

Michelle Valente, Cyril Joly, Arnaud de La Fortelle

The paper proposes a CNN-based framework that fuses 2D laser scanner and monocular camera data to estimate odometry in real time without requiring sensor calibration.

#sensor fusion#odometry estimation#convolutional neural networks#real-time processing

eess.AS2019

Fast and Robust 3-D Sound Source Localization with DSVD-PHAT

Francois Grondin, James Glass

The paper proposes Difference SVD-PHAT (DSVD-PHAT), a variant of SVD-PHAT for 3‑D sound source localization that remains robust in noisy environments while reducing computational l…

#sound source localization#robotics#singular value decomposition#phase transform

cs.GR2019

Coherent Point Drift Networks: Unsupervised Learning of Non-Rigid Point Set Registration

Lingjing Wang, Xiang Li, Jianchun Chen +1

The paper introduces CPD-Net, an unsupervised deep learning framework that learns a continuous displacement field to perform real-time non-rigid point set registration without iter…

#point set registration#non-rigid transformation#deep learning#unsupervised learning

cs.SI2019

Real-time Event Detection on Social Data Streams

Mateusz Fedoryszak, Brent Frederick, Vijay Rajaram +1

The paper presents a real-time system that detects events by clustering trending entities in high‑volume social media streams such as Twitter, and introduces evaluation datasets an…

#event detection#streaming clustering#social media analytics#real-time processing