This opportunity is based in Lausanne

Internship - Object Detection in Text Documents

Apply

Business documents are challenging to process as most of them contain graphical objects such as tables, images, logos, schemas or signatures. These objects might not only be rich in information, but their presence can also degrade the performance of open-source optical character recognition tools (OCR). Document object detection (DOD) is therefore a crucial step in processing pipelines for digitized documents. It allows to extract each graphical object to be further processed with a dedicated method, such as handwritten signature verification or table parsing.

Despite recent deep learning approaches, DOD remains a challenging computer vision task, as each type of graphical object can appear in a wide range of layouts, shapes or textures. In addition, a very high accuracy is required in order to enhance downstream applications such as document classification or information extraction.

In this role

In this internship, you will explore state-of-the-art document object detection approaches and eventually implement and deploy a web-based prototype of your solution.

You will also study the automatic verification of signatures, which is about determining first whether there is a signature and then that the detected signature matches known samples. This is an important problem to solve, since handwritten signatures are still a widely accepted way of authentication in the business world today.

 

The goal of the internship is to:

  • Study and compare state-of-the-art machine learning methods for document object detection and for signature verification. The student will have the opportunity to deepen their knowledge in these two fields.
  • Design and implement a tool for document object detection and signature verification. The intern will be able to sharpen their machine learning and software development skills.

What we offer

Join our team as intern and you will find a young, dynamic and culturally diverse working environment.

About your profile

  • Interest and strong knowledge in Machine Learning and Computer Vision
  • Experience with Python (NumPy stack, common ML and computer vision libraries)
  • Experience with Git, open-source OCR libraries, Docker and REST/microservices is a plus

If you are INTERESTED in applying for this position, please send us your complete application (CV, cover letter, letter of reference, diplomas and certificates).

By continuing to browse this site, you accept the use of cookies or similar technologies whose purpose is to produce statistics on visits to our site (tests and measurement of visitor numbers, visit frequency, page views and performance) and to offer you content and promotions which will be of interest to you.

Our cookie policy has been updated. Feel free to manage your preferences.

close
save

Manage your cookie preferences

Update your cookie preferences

Find out about the type of cookies stored on your device, accept or block them for the entire site, all services or on a service-by-service basis.

OK, accept all

Visitor flow

These cookies provide us with insight into traffic sources and allow us to better understand our visitors anonymously.

(Google Analytics and CrazyEgg)

New

Sharing tool

Social media cookies allow content sharing on your preferred networks.

(ShareThis)

New

Visitor understanding

These cookies are used to track visitors across websites.

The intention is to enable us to offer more relevant, targeted content to existing contacts (ClickDimensions) and display ads that are relevant and engaging for users (Facebook Pixels).

 

New
For more information about these cookies and our cookie policy, click here