Technical Management For Machine Learning & Backend Development On OCR Internship in Mumbai at IIT Bombay
Technical Management For Machine Learning & Backend Development On OCR
Start Date
1 Nov - 30 Nov' 20
Duration
6 Months
Stipend
2000-4000 /month
Apply By
31 Oct' 20
252 applicants
The hiring for this internship will be online and the company will provide work from home/ deferred joining till current COVID-19 situation improves
Internship
Internship
252 applicants
About IIT Bombay
The Indian Institute of Technology, Bombay (IITB) is one of the fifteen higher institutes of technology in the country set up with the objective of making facilities available for higher education, research, and training in various fields of science and technology. With the same mission and vision, Prof. Ganesh Ramakrishnan is gearing to take rural India a leap ahead. For his outstanding contributions, he has also been awarded the IBM Faculty Award in 2011. IIT Bombay has also honored Prof Ganesh's work on "Adaptive framework for end-to-end corrections in Indic OCR".
About the internship
Selected intern's day-to-day responsibilities include:

1. Understanding the requirements in detail and eliciting them from the involved stakeholders
2. Leading the machine learning team and guiding them in the challenges they are facing
3. Collaborating with the backend development team and ensuring the requirements are implemented correctly and the project is headed on the right track
4. Organizing meetings with all the required stakeholders and discussing the requirements, issues, challenges, updates, and progress relevant to the project
5. Overseeing that the development and testing are done as per industry standards and a proper pipeline is followed to ensure smooth operations of the project
6. Deploying the system on private servers and then handling the server operations
7. Managing the team and overseeing the tasks done by them
8. Recruiting skilled team members as per the needs of the project
9. Co-ordinating with industry experts and taking technical inputs from them in the challenges faced in the project
10. Maintaining the product backlog and software documentation
Skill(s) required
Python Software Testing Django Machine Learning C++ Programming
Learn these skills on Internshala Trainings
Who can apply

Only those candidates can apply who:

1. are available for full time (in-office) internship

2. can start the internship between 1st Nov'20 and 30th Nov'20

3. are available for duration of 6 months

4. have relevant skills and interests

Other requirements

1. Must have done several projects in backend development/web development and also strong knowledge in machine learning

2. Must have excellent communication skills in the English language

3. Must have good people management, collaboration, and leadership skills

4. Must have a strong computer science technical background with awareness about machine learning, servers, databases, API, and software testing

5. Must have basic knowledge in Python or C++

6. Should have done multiple projects in machine learning

7. Must have hands-on experience in coding in software development/backend projects in technologies like Java/C++/Python/Django, etc.

8. Must have graduated in 2020 (candidates passing out in 2021 can also be considered if they have their final semester available for the internship and have no course work commitments)

9. Must be available full time for the internship (minimum 9 hours per day)

10. Must be available for a minimum duration of 6 months (candidates available for a year are highly preferred)

11. Must be willing to take leadership and initiative to take the project to new heights

Perks
Certificate Letter of recommendation Flexible work hours 5 days a week
Additional Information

We are working on detecting the layout of the OCRed text, preserve it, and reflect it in the to automate the annotation task. Optical character recognition (OCR) is the process of converting the document images into an editable electronic format. This has many advantages like data compression, enabling search or edit options in the images/text, and creating the database for other applications like machine translation, speech recognition, and enhancing dictionaries and language models. OCR in Indian languages is quite challenging due to richness in inflections. Using open-source and commercial OCR systems, we have observed the word error rates (WER) of around 20-50% on printed documents in four different Indic languages. Moreover, developing a highly accurate OCR system with accuracy as high as 90% is not useful unless aided by the mechanism to identify errors. So, we started with the problem of developing "OpenOCRCorrect", an end-to-end framework for error detection and corrections in Indic-OCR. Our models outperform state-of-the-art results in 'Error Detection in Indic-OCR' for six Indic languages with varied inflections and we have solved the out of vocabulary problem for “Error Correction in Indic-OCR” in our ICDAR-2017 conference paper. We further improve the results with the help of sub-word embeddings in our ICDAR-2019 conference paper.

The demo video for our framework is https://www.youtube.com/watch?v=u9bqUDrGugc.
To install the software, you can go to https://github.com/rohitsaluja22/OpenOCRCorrect and follow the instructions given in https://www.youtube.com/watch?v=0hcdlF-zn8E. There is an immediate demand to keep the softcopy of the Indian preserved texts. Currently, we are targeting Sanskrit. Although the OCR tools available online do a decent job on English texts, they are not optimized for Indic languages. Thus, we developed an in-house OCR model for the same. The model can detect text with maximum level accuracy and can draw bounding boxes on each line of the text. Further, in the digitization process of such texts, the second step would be spelling correction and formatting of the text detected by the OCR models.

'ICDAR 2019 Post-OCR competition': Our team "CLAM" secured 2nd position in the multilingual PostOCR competition at ICDAR'19. Our model achieved the highest corrections of 44% in Finnish, which is significantly higher than the overall topper (8% in Finnish).
Our final report for the same - (https://drive.google.com/file/d/15mxNOM9PiXBnffi7MOa8wUw33nj1xBp/view?usp=drive_open) and poster available (https://drive.google.com/file/d/1uuBWu1LQ1QZ49SCgLBoB1er4HpWSzmcx/view).

'ICDAR2019': You can read the paper here -https://www.cse.iitb.ac.in/~rohitsaluja/PID6011473.pdf
'ICDAR2017': You can read the paper here - https://ieeexplore.ieee.org/document/8269944

'ICDAR-OST 2017':
a. OpenOCRCorrect, you can read the paper here - https://ieeexplore.ieee.org/abstract/document/8270254
b. The source code for our framework is available here - https://github.com/rohitsaluja22/OpenOCRCorrect

Candidates need to understand that this is a technical management role required to lead a backend team working in C++/Python and also lead the ML Team. Technical Guidance will be provided by industry experts or professors but the candidate needs to execute the work with the team and manage efficiently.

Number of openings
1

Save yourself from fraud!

If an employer asks you to pay any security deposit, registration fee, laptop fee, etc., do not pay and notify us immediately. Remember, Internshala doesn't charge a fee from the students to apply to a job or an internship & we don't allow other companies to do so either.