Charalambos Poullis

Charalambos Poullis

Latest — Sep 9, 2022

Simpler is better: Multilevel Abstraction with Graph Convolutional Recurrent Neural Network Cells for Traffic Prediction

The preprint of our paper Simpler is better: Multilevel Abstraction with Graph Convolutional Recurrent Neural Network Cells for Traffic Prediction is available on arxiv.org. The work is co-authored by Naghmeh Shafiee Roudbari, Zachary Patterson, Ursula Eicker, and Charalambos Poullis. TL;DR: We present a sequence-to-sequence architecture to extract the

More issues

Unsupervised Structure-Consistent Image-to-Image Translation

The preprint of our paper Unsupervised Structure-Consistent Image-to-Image Translation is available on arxiv.org. The work is co-authored by Shima Shahfar and Charalambos Poullis. TL;DR: We present a structure-consistent image-to-image translation technique based on the swapping autoencoder. We introduce an auxiliary module which forces the generator to learn to

Motion Estimation for Large Displacements and Deformations

The preprint of our paper Motion Estimation for Large Displacements and Deformations is available on arxiv.org. The work is co-authored by Qiao Chen and Charalambos Poullis. TL;DR: Large-scale images such as wide-area motion imagery cannot be handled by deep learning architectures. The standard procedural method for image-based reconstruction

Predicting Human Performance in Vertical Hierarchical Menu Selection in Immersive AR Using Hand-gesture and Head-gaze

The paper titled Predicting Human Performance in Vertical Hierarchical Menu Selection in Immersive AR Using Hand-gesture and Head-gaze authored by Majid Pourmemar, Yashas Joshi and Charalambos Poullis will be presented at the 15th Conference on Human System Interaction, 2022. TL;DR: User studies are expensive, time-consuming, and primarily based on

SaccadeNet: Towards Real-time Saccade Prediction for Virtual Reality Infinite Walking

The preprint of our paper SaccadeNet: Towards Real-time Saccade Prediction for Virtual Reality Infinite Walking is available on arxiv.org. The work is co-authored by Yashas Joshi and Charalambos Poullis. TL;DR: We propose a deep neural network for predicting saccades in VR applications of moderate cognitive workload. During saccades,

Tractable Large-Scale Deep Reinforcement Learning

Some results from the latest paper Tractable Large-Scale Deep Reinforcement Learning by Nima Sarang and Charalambos Poullis, on deep reinforcement learning. The code is publicly available at https://github.com/nsarang/road-extraction-rl. TL;DR: We reformulate the problem of road extraction by training an RL agent to traverse the roads

Adaptive Memory Management for Video Object Segmentation

The paper titled Adaptive Memory Management for Video Object Segmentation authored by Ali Pourganjalikhan and Charalambos Poullis will be presented at the 19th Conference of Robots and Vision (CRV) 2022. TL;DR: Given a sequence of frames and the masks of the target objects in their first appearance, the goal

Managing real-world and virtual motion

A US Patent application was published for the technology presented in the IEEE Access journal and the ACM SIGGRAPH poster on redirected walking by Yashas Joshi and Charalambos Poullis. TL;DR: Navigating large-scale virtual environments in room-scale physical spaces using natural locomotion is problematic since users will most certainly encounter

Multi-view Gradient Consistency for SVBRDF Estimation of Complex Scenes under Natural Illumination

Some results from the technique detailed in the paper "Multi-view Gradient Consistency for SVBRDF Estimation of Complex Scenes under Natural Illumination" by Alen Joy and Charalambos Poullis. TL;DR: The diffuse component is independent of the viewer's direction, therefore, when solving for BRDFs, the gradients from multi-view images must be