Zum Hauptinhalt springen

Operator Actions Recognition using Deep Learning

Aktuellste Jobs

Operator Actions Recognition using Deep Learning

BE-2020-070KUL

WORK OFFER Ref. No. BE-2020-070KUL

Employer Information

Employer: Flanders Make
HR

Website: www.flandersmake.be

Gaston Geenslaan 8 - 3001 Leuven - BELGIUM
Oude Diestersebaan 133 – 3920 Lommel - BELGIUM
3001 Leuven
Belgium

Location of placement: Leuven, Belgium
Nearest airport: Brussels
Working hours per week: 38.0
Working hours per day: 7.6

Number of employees: 100
Business or products: research for the manufacturing industry

Student Required

General Discipline: 11-COMPUTER AND INFORMATION SCIENCES
14C-ELECTRICAL AND ELECTRONICS
ENGINEERING

Completed years of study: 3

Field of Study:
or related field

Student status requirements: Student status during the entire internship is
mandatory: please include Certificate of
Enrolment with nomination.

Language required: English Excellent

Required Knowledge and Experiences: Other requirements:

• Bachelor degree;

• Knowledge of image processing, computer vision, machine learning is
highly recommended;

• Good programming skills in Python or C++

• Experience in open-source deep learning frameworks such as TensorFlow
or PyTorch preferred


• Passionate by research and new technologies with focus on applications
that includes machine learning, deep learning and computer vision

• Result oriented, responsible and proactive;

• Eager to learn and a team player.

Only students with EEA or Swiss nationality!

Work Offered

Operator Actions Recognition using Deep Learning

The detection and recognition of operator actions from a data streams is nowadays a popular challenge, with the potential to aid in operator fast training,
monitoring and fault detection. In this internship the aim is to study the state of the art available techniques that address this challenge of interpreting
operator actions in industrial environment.
The expected outcome is a real-time operator action detection and action recognition system.

This internship is linked to the FAMAR project, in which the overall goal is to create an economically feasible user-centred Augmented Reality application
methodology for flexible assembly and inspection operations in a low volume/high mix manufacturing environment.

In this context, the goal of this internship is to perform a state of the art study of the latest research and development done in the area of operator actions
recognition, oriented toward industrial applications in a controlled environment. A list of predefined actions such as: caulking, hammering and/or screwing
will be selected.
Multiple vision sensors will be made available during the internship (2D and 3D) in order to validate. During this internship, after the state of art study you
will have to collect and annotate data of pre-selected actions to be recognized. Then, implement, apply and benchmark the selected approaches to perform
operator action recognition on the collected dataset.

Number of weeks offered: 16 - 26 Working environment: Research and development
Within the months: 06-APR-2020 - 18-DEC-2020 Gross pay: 200 EUR / Week
Or within: - Deduction to be expected: 0
Company closed within: - Payment method / time of first

payment:
/

Latest possible start date:

Accomodation

Canteen at work: No
Expected type of accommodation: Student dormitory Estimated cost of lodging: 100 EUR / Week
Accommodation will be arranged by: IAESTE Estimated cost of living incl. lodging: 200 EUR / Week

Additional Information

Nomination Information

Deadline for nomination: 15-MAR-2020

Date: 27-FEB-2020 On behalf of receiving country: Annelies Vermeir

IAESTE BELGIUM

Operator Actions Recognition using Deep Learning

Löwen
Praktikum
ab April
Bachelor

Veröffentlicht am 09.04.2020

Jetzt Job teilen