Postgraduate – Page 2 – Postgraduates by Research

PGRs meeting and Research Presentations – April 2016

Amr Ahmed / April 14, 2016April 14, 2016 / Postgraduate, Uncategorized

The monthly PGRs Research Presentations was held on Wed. 13^th April, 2pm, Room MC3108.

This session we had the following presentations:

PGRs Monthly meeting_April2016 (Slides ) Agenda Speaker –> A quick look at the new “PGRs Management System”, PGRs Blog. Discussion of the activities plan. Update and plan for the “Showcase Event” Announcements, AOB, & closing		Title: Life-long Spatio-temporal Exploration of Dynamic Environments: An overview.
		By: Joao Santos
		Abstract: The primary purpose of robotic exploration is to autonomously acquire a complete and precise model of the robot’s operational environment. To explore efficiently, the robot has to direct its attention to environment areas that are currently unknown. If the world was static, these areas would simply correspond to previously unvisited locations. In the case of dynamic environments, visiting all locations only once is not enough, because they may change over time. Thus, dynamic exploration requires that the environment locations are revisited and their (re-) observations are used to update a dynamic environment model. However, revisiting the individual locations with the same frequency and on a regular basis is not efficient because the environment dynamics will, in general, not be homegeneous, (i.e. certain areas change more often and the changes occur only at certain times). Similarly to the static environment exploration, the robot should revisit only the areas whose states are unknown at the time of the planned visits. Thus, the robot has to use its environment model to predict the uncertainty of the individual locations over time and use these predictions to plan observations that from a theoretical point of view improve its knowledge about the world’s dynamics. Hence, the observations are scheduled in order to obtain information about the environment changes, which are mainly caused by human-activity. As a consequence, using schedules motivated by the changes in metric maps increases the chance to extract dynamics that are essential for object learning and activity recording tasks.

Farewell and Thank You

Amr Ahmed / March 13, 2016April 13, 2016 / Postgraduate

At the end of the March’s meeting, we had a Farewell and Thank You for previous “PGRs Students Reps” and PGRs who have been helping in various activities, including Reading Group, trips, coffee mornings, and the showcase.

PGRs Representatives:

–1st Students Rep: Touseef Quraishi

–2nd Students Reps:

Christian
Mohammadreza.

PGRs who have been helping:

(George; showcase events & reading group – Saddam; Showcase Events & support PGRs meeting – Francesco; trips & showcase – Ibrahim; trips & showcase – Christian; regular presentations & reading group – Talal; Reading group – Hussein; showcase)

Thanks to all, and best wishes for the years to come.

PGRs meeting and Research Presentations – Nov. 2015

Amr Ahmed / November 11, 2015 / Postgraduate

The monthly PGRs Research Presentations was held on Wed. 11^th November, 2pm, Room MC3108.

This session we had the following presentations:

Title: “Profiling User Engagement with Promotional Social Media Content“.		Title: “A Bio-inspired Collision Detection Vision System Embedded for Autonomous Micro-robots ”
By: Jamie Mahoney		By: Cheng Hu
Abstract: Organisations, retailers and brands have a long-established need to gain insight into the characteristics of their customers, users and online followers. Using Twitter as a case study, we describe a method of creating engagement profiles of users based on qualitative analyses of the social media content with which they have publicly engaged. By clustering these engagement profiles, we extend previous work by not only showing how people within a social graph can be clustered in useful ways, but also how these users are likely to engage with specific types of social media content – thus allowing for the creation of targeted social media content strategies. Our findings demonstrate that ‘traditional’ methods of social graph segmentation do not reflect groupings of similar users in terms of engagement behaviour, we also demonstrate that user engagement behaviours do not vary dramatically over time. We provide suggestions for how these findings might be used in the creation of effective strategies for companies, and organisations, wishing to issue promotional material via social media platforms.		Abstract: The vision system takes inspiration from locusts’ response in detecting fast approaching objects. Neurophysiological research suggested that locusts use a wide-field visual neuron called lobula giant movement detector (LGMD) to respond to imminent collisions. In this work, we present the implementation of selected neuron model by a low-cost ARM processor as part of a composite vision module. The developed system is able to perform image acquisition and processing independently. The vision module is placed on top of a micro-robot to initiate obstacle avoidance behaviour. Both simulation and real-world experiments were carried out to test the reliability and robustness of the vision system. The results of the performed experiments with different scenarios demonstrated the amenability of the developed bio-inspired vision system to be used as a low-cost embedded module in autonomous robots with high precision.


Then our usual catch-up agenda:

StatsDay (Statistics Session)

Amr Ahmed / October 30, 2015November 11, 2015 / Postgraduate

On 28th October 2015, LSoCS PGRs had a StatsDay event, thanks to Dr Phil Assheton for being with us.

The event covered main concepts around the statistics and how it helps in experiments and evaluations, including parametric vs non-parametric techniques, the interpretation of the p-value and what conclusion we can take.

The event also included a hands-on practical part, where participants worked some prepared examples and programmed in R.

Thanks to all involved.

PGRs meeting and Research Presentations – Oct. 2015

Amr Ahmed / October 20, 2015November 11, 2015 / Postgraduate, Uncategorized

The monthly PGRs Research Presentations was held on Wed. 14^th October, 2pm, Room MC3108.

This session we had the following presentations:

Title: “Modelling LGMD2 Visual Neuron System“.		Title: “Compressed video matching: Frame-to-frame revisited”
By: Qinbing Fu		By: Saddam Bekhet
Abstract: Two Lobula Giant Movement Detectors (LGMDs) have been identified in the lobula region of the locust visual system: LGMD1 and LGMD2. LGMD1 had been successfully used in robot navigation to avoid impending collision. LGMD2 also responds to looming stimuli in depth, and shares most the same properties with LGMD1; however, LGMD2 has its specific collision selective responds when dealing with different visual stimulus. Therefore, in this paper, we propose a novel way to model LGMD2, in order to emulate its predicted bio-functions, moreover, to solve some defects of previous LGMD1 computational models. The mechanism of ON and OFF cells, as well as bio-inspired nonlinear functions, are introduced in our model, to achieve LGMD2’s collision selectivity. Our model has been tested by a miniature mobile robot in real time. The results suggested this model has an ideal performance in both software and hardware for collision recognition.		Abstract: This presentation is about an improved frame-to-frame (F-2-F) compressed video matching technique based on local features extracted from reduced size images, in contrast with previous F-2-F techniques that utilized global features extracted from full size frames. The revised technique addresses both accuracy and computational cost issues of the traditional F-2-F approach. Accuracy is improved through using local features, while computational cost issue is addressed through extracting those local features from reduced size images. For compressed videos, the DC-image sequence, without full decompression, is used. Utilizing such small size images (DC-images) as a base for the proposed work is important, as it pushes the traditional F-2-F from off-line to real-time operational mode. The proposed technique involves addressing an important problem: namely the extraction of enough local features from such a small size images to achieve robust matching. The relevant arguments and supporting evidences for the proposed technique are presented. Experimental results and evaluation, on multiple challenging datasets, show considerable computational time improvements for the proposed technique accompanied by a comparable or higher accuracy than state-of-the-art related techniques.


Then our usual catch-up agenda:		PGR Month (as per the GS email). Reminder of the StatsDay session (28^th Oct, Lab B). Expected Deadline (for Progress Panel) v.early Nov. Announce @ “App Fest”: Required ~5 supervisors (PGR/MComp):