Search Results - "Liwicki, Stephan"

Orientation-Aware Semantic Segmentation on Icosahedron Spheres by Zhang, Chao, Liwicki, Stephan, Smith, William, Cipolla, Roberto

“…We address semantic segmentation on omnidirectional images, to leverage a holistic understanding of the surrounding scene for applications like autonomous…”
Get full text

Euler Principal Component Analysis by Liwicki, Stephan, Tzimiropoulos, Georgios, Zafeiriou, Stefanos, Pantic, Maja

“…Principal Component Analysis (PCA) is perhaps the most prominent learning tool for dimensionality reduction in pattern recognition and computer vision…”
Get full text

HexNet: An Orientation-Aware Deep Learning Framework for Omni-Directional Input by Zhang, Chao, Liwicki, Stephan, He, Sen, Smith, William, Cipolla, Roberto

“…While omni-directional sensors provide holistic representations typical deep learning frameworks reduce the benefits by introducing distortions and…”
Get full text

Online Kernel Slow Feature Analysis for Temporal Video Segmentation and Tracking by Liwicki, Stephan, Zafeiriou, Stefanos P., Pantic, Maja

“…Slow feature analysis (SFA) is a dimensionality reduction technique which has been linked to how visual brain cells work. In recent years, the SFA was adopted…”
Get full text

Efficient Online Subspace Learning With an Indefinite Kernel for Visual Tracking and Recognition by Liwicki, S., Zafeiriou, S., Tzimiropoulos, G., Pantic, M.

“…We propose an exact framework for online learning with a family of indefinite (not positive) kernels. As we study the case of nonpositive kernels, we first…”
Get full text

Embodied Visual Navigation With Automatic Curriculum Learning in Real Environments by Morad, Steven D., Mecca, Roberto, Poudel, Rudra P. K., Liwicki, Stephan, Cipolla, Roberto

“…We present NavACL, a method of automatic curriculum learning tailored to the navigation task. NavACL is simple to train and efficiently selects relevant tasks…”
Get full text

CoMBiNED: Multi-Constrained Model Based Planning for Navigation in Dynamic Environments by Pandya, Harit, Poudel, Rudra P.K., Liwicki, Stephan

“…Recent deep reinforcement learning (DRL) approaches have achieved high success rate in map-less dynamic obstacle avoidance tasks. However, navigation in unseen…”
Get full text

Euler Principal Component Analysis: Human-Computer Interaction: Real-Time Vision Aspects of Natural User Interfaces by LIWICKI, Stephan, TZIMIROPOULOS, Georgios, ZAFEIRIOU, Stefanos, PANTIC, Maja

Get full text

Automatic recognition of fingerspelled words in British Sign Language by Liwicki, Stephan, Everingham, Mark

“…We investigate the problem of recognizing words from video, fingerspelled using the British Sign Language (BSL) fingerspelling alphabet. This is a challenging…”
Get full text

Robust online subspace learning by Liwicki, Stephan

“…In this thesis, I aim to advance the theories of online non-linear subspace learning through the development of strategies which are both efficient and robust…”
Get full text

DiaLoc: An Iterative Approach to Embodied Dialog Localization by Zhang, Chao, Li, Mohan, Budvytis, Ignas, Liwicki, Stephan

“…Multimodal learning has advanced the performance for many vision-language tasks. However, most existing works in embodied dialog research focus on navigation…”
Get full text

ReCoRe: Regularized Contrastive Representation Learning of World Model by Poudel, Rudra P.K., Pandya, Harit, Liwicki, Stephan, Cipolla, Roberto

“…While recent model-free Reinforcement Learning (RL) methods have demonstrated human-level effectiveness in gaming environments, their success in everyday tasks…”
Get full text

DiaLoc: An Iterative Approach to Embodied Dialog Localization by Zhang, Chao, Li, Mohan, Budvytis, Ignas, Liwicki, Stephan

“…Multimodal learning has advanced the performance for many vision-language tasks. However, most existing works in embodied dialog research focus on navigation…”
Get full text

Self-Supervised Consistent Quantization for Fully Unsupervised Image Retrieval by Wu, Guile, Zhang, Chao, Liwicki, Stephan

“…Unsupervised image retrieval aims to learn an efficient retrieval system without expensive data annotations, but most existing methods rely heavily on…”
Get full text

ReCoRe: Regularized Contrastive Representation Learning of World Model by Poudel, Rudra P. K, Pandya, Harit, Liwicki, Stephan, Cipolla, Roberto

“…While recent model-free Reinforcement Learning (RL) methods have demonstrated human-level effectiveness in gaming environments, their success in everyday tasks…”
Get full text

Reinforcement Learning with Fast and Forgetful Memory by Morad, Steven, Kortvelesy, Ryan, Liwicki, Stephan, Prorok, Amanda

“…Nearly all real world tasks are inherently partially observable, necessitating the use of memory in Reinforcement Learning (RL). Most model-free approaches…”
Get full text

POPGym: Benchmarking Partially Observable Reinforcement Learning by Morad, Steven, Kortvelesy, Ryan, Bettini, Matteo, Liwicki, Stephan, Prorok, Amanda

“…Real world applications of Reinforcement Learning (RL) are often partially observable, thus requiring memory. Despite this, partial observability is still…”
Get full text

Recurrent Reinforcement Learning with Memoroids by Morad, Steven, Lu, Chris, Kortvelesy, Ryan, Liwicki, Stephan, Foerster, Jakob, Prorok, Amanda

“…Memory models such as Recurrent Neural Networks (RNNs) and Transformers address Partially Observable Markov Decision Processes (POMDPs) by mapping trajectories…”
Get full text

Lifted Semantic Graph Embedding for Omnidirectional Place Recognition by Zhang, Chao, Budvytis, Ignas, Liwicki, Stephan, Cipolla, Roberto

“…Typical place recognition is dependent on the visual appearance and camera position of query images, without explicit use of domain knowledge and geometric…”
Get full text

Full-Angle Quaternions for Robustly Matching Vectors of 3D Rotations by Liwicki, Stephan, Minh-Tri Pham, Zafeiriou, Stefanos, Pantic, Maja, Stenger, Bjorn

“…In this paper we introduce a new distance for robustly matching vectors of 3D rotations. A special representation of 3D rotations, which we coin full-angle…”
Get full text