Computer Vision and Pattern Recognition Experts

It-Jim is a group of experts with PhD degrees in various mathematical disciplines. We conduct the high-quality research in the fields of computer vision, pattern recognition, machine learning, artificial intelligence, augmented reality, signal and image processing. We combine a deep understanding of the signal and image analysis theory with advanced programming skills. We are eager to work on challenging projects and solve complicated and interesting problems.

 
$5,000+
 
$50 - $99 / hr
 
10 - 49
 Founded
2015
Show all +
Kharkiv, Ukraine
headquarters

Portfolio

Key clients: 
Esticom, VisionLab, Foresight automotive, Geomagical labs
VISUAL INSPECTION  OF ARTIFICIAL HEART VALVE Image

VISUAL INSPECTION OF ARTIFICIAL HEART VALVE

When you replace a heart valve with the artificial one (which is now a common thing in modern medicine), it is crucial to be sure you place a good substitution. Implantation of a valve introduces some serious heart risks and pathological phenomena, so effective and adequate inspection of the artificial valve becomes a very important issue.

In this project, we have developed a computer vision-based tool for

visual inspection of artificial valves at the end stage of manufacturing. It facilitates valve condition assessment and provides some handful tools for the acceleration of manual testing. As an input, the program uses videos with the recorded operating heart valve. With the built-in custom video player, the program helps to analyze the condition of the prosthesis and check if the valve meets all inspection criteria. The computer vision algorithms are used to detect valve buckles and leaflets as well as in the segmentation of the central jet area. The features implemented in the program include but are not limited to zoom, screen capture, grid, unit calibration, ruler, central hole (visualizes the segmentation of the valve’s central hole), and geometric orifice area (visualizes the detected leaflets).

KEY POINTS:

  • Automatic control of artificial heart valve operation
  • Custom solution for real-time segmentation of central jet area
  • The efficient fusion of classic computer vision and deep neural nets for the handling of challenging illumination conditions
MULTI-PLAYER SPORT TRACKER Image

MULTI-PLAYER SPORT TRACKER

Computer vision has a broad list of interesting applications in sports. In particular, multiple player tracking, ball tracking, action recognition, automatic goal-scoring, view generation and many more.

We have built a large-scale system for real-time sports analytics. The main goal of the project was to replace the human operator and automatically find the most interesting area on the hockey/football/

basketball pitch. The region-of-interest (ROI) control algorithm required pitch mask segmentation for avoiding false detections of the player blobs, panorama stitching from multiple cameras in real time, adaptive filtering and thresholding to detect the player candidates and custom tracker for robust keeping the traces of players. A specific clustering procedure was developed in order to build a proper logic for selection and smooth control of the ROI.

The key advantage of the created tracker is extremely high performance. It works at 120 FPS and multiple system instances can be simultaneously installed in the cloud. This makes the solution quite cheap and flexible, ROI tracking and streaming of the outputs are performed for many clients all over the globe on a real-time basis.

KEY POINTS:

  • Multiple objects are tracked on a real-time basis
  • The tracker can combine results from multiple cameras
  • The solution provides data for sport analytics
  • Region of interest for live broadcasting can be extracted automatically
DOCUMENT ANALYSIS AND RECOGNITION Image

DOCUMENT ANALYSIS AND RECOGNITION

Automatic document analysis is a key part of the overall document recognition process. A common scenario is when the user takes a picture by mobile phone or tablet and the goal is to automatically parse and recognize content from the captured document. Such as pictures, tables, text data, links, etc. There are several challenges in this case: geometric distortions of the paper, varying illumination, occlusions.

Nevertheless, we have built the page unwrapping engine, which successfully handles the above problems. The developed algorithm contains several key components: Preprocessor, Feature Extractor, Geometric Model Estimator and Refiner. The preprocessor performs image filtering, roughly locates the document boundaries and extracts vertical and horizontal spans (lines and text regions). Feature Extractor module performs parsing of document content. For robustness, we have fused several types of features including corners and edges from images preprocessing in different ways. The output nonlinear grid of points is used as an input for the next system component: Geometric Model Estimator. In order to properly account for the nonlinear geometric distortions of the document, we have constructed a specific 2D-3D model, where page shape is reconstructed as a 3D surface together with the 6 DoF camera position in 3D space.Finally, after the estimation of model parameters, the document is dewarped and most of the existing distortions are corrected. Developed dewarping engine works very accurately and performs distortion correction within a reasonable time frame.

KEY POINTS:

  • Handling illumination changes
  • Correction of nonlinear geometric distortions
  • Automatic page border detection
  • 2-step procedure returns well-refined document layout
SYMBOL RECOGNITION Image

SYMBOL RECOGNITION

Pattern recognition is an extensive field of research since there are a number of things to detect and recognize from the real world.

In this project, our goal was to develop an engine for automatic counting of the various symbols on scanned building plans. Our custom algorithm was able to handle hundreds of different symbols, including complex ones consisting of multiple small sub-patterns and text. It

is robust to rotation, scaling and partial symbol occlusions. Develop engine is based on a purely classic computer vision algorithm, works very fast and does not require GPU, which makes it very cheap to use in the cloud. In addition, a comparative test with market competitors indicates an advantage from both accuracy and performance points of view.

KEY POINTS:

  • Automatic recognition of symbols with different complexity
  • Optimized search using multi-threading
  • Custom decomposition algorithm
  • Rotation and scale-invariant recognition engine
  • Successfully integrated into an engineering platform
PLANAR TRACKER FOR MOBILE AR Image

PLANAR TRACKER FOR MOBILE AR

Marker-based augmented reality (AR) is extremely popular nowadays. However, full user immersion is possible only in the case of robust real-time computer vision solutions working on the mobile device. We have developed a custom hybrid tracking system based on local feature tracking and template-based matching. The tracking engine tracks the homography changes using an optical flow algorithm and then refines the

residual warp using the optimized template matcher. Developed tracker is robust to extremal angles, drift-free and very accurate. Algorithms work at 60 FPS on Android and iOS.

KEY POINTS:

  • Custom tracker is resistant to extremal angles and drift
  • 60 FPS on mobile device (iOS+Android)
  • Fusion of feature-based and template-based trackers provides high robustness
  • Successfully used in custom mobile AR SDK
3D-INSPECTION BY DRONES Image

3D-INSPECTION BY DRONES

Drones are widely used for various tasks including visual inspection, facades and buildings reconstruction, surveillance and tracking, object detection and recognition, etc.

In this project, we have built a web-tool for visual inspection of buildings. The core technology was a customized 3D reconstruction pipeline working with both image sequences or videos from the drone camera. We have implemented

structure-from-motion (SfM), point cloud densification, meshing and texturing modules. The resulting pipeline was integrated into the web-service, where the user was able to load the data for reconstruction, analyze the resulting 3d model, view the fine details and save the results for further steps.

KEY POINTS:

  • 3D point cloud reconstruction engine is implemented
  • Successfully integrated into web-service
  • Full pipeline including SfM, meshing and texturing is created
MOBILE IMAGE RETRIEVAL Image

MOBILE IMAGE RETRIEVAL

Visual image search and content-based image retrieval (CBIR) is a very hot topic in computer vision and machine learning communities. It is a core of many systems in retail, entertainment, biometrics, robotics, and other domains. The main technical challenge is to be able to create a reliable solution capable to retrieve correct results from millions of images. Moreover, such a system should work in real-time and, ideally,

without internet connection on mobile devices or tablets. In our project, we have successfully developed a mobile image retrieval system for augmented reality applications. For this purpose, we have optimized and customized the bag-of-words (BoW) algorithm using a properly constructed hierarchical tree for the features vocabulary. As a result, local features from the test images were matched extremely fast. Our CBIR system worked on a mobile device without an internet connection and returned the visual search results within only 2 milliseconds.

KEY POINTS:

  • Instant retrieval offline on a mobile device
  • 2 ms time for 1000 images database
  • Works on iOS and Android
  • Integrated into custom mobile AR SDK

Reviews

Sort by

Development for Cloud-Based AR Solution

"They’re experts in their field."

Quality: 
5.0
Schedule: 
5.0
Cost: 
5.0
Willing to refer: 
5.0
The Project
 
$10,000 to $49,999
 
Nov 2019 - Ongoing
Project summary: 

IT-Jim provided research and consulting services on computer vision and augmented reality technology. They helped the client develop a demo of a cloud-based web application. 

The Reviewer
 
1-10 Employees
 
Ontario, Canada
James Babarinde
CEO, NovaStack
 
Verified
The Review
Feedback summary: 

As a result of IT-Jim’s work, the client was able to make gains with the newest version of their demo. Communication was fluid and the team has a depth of expertise in the computer vision field. They delivered within stated timeframes and provided good quality work. 

A Clutch analyst personally interviewed this client over the phone. Below is an edited transcript.

BACKGROUND

Introduce your business and what you do there.

I’m the CEO of a boutique agency. We recently moved to expand our augmented reality services by creating proprietary software. 

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with IT-Jim?

We had architected a solution that was rooted in web-based augmented reality and we needed a development team to build it out and advise us on the project. 

SOLUTION

What was the scope of their involvement?

There was a heavy research phase. There were several ideas we wanted to try and they helped us determine which areas of our solution would be worth pursuing. They used a mix of coding languages including C++ and JavaScript to build us a cloud-based solution. 

We provided 3D models and some assets to demonstrate the computer vision algorithm since they were building a utility solution. 

What is the team composition?

My main point of contact was Ievgen (CEO & Founder). There were also several developers that I worked with. 

How did you come to work with IT-Jim?

I found them online. I chose them because of their prior experience in computer vision. 

How much have you invested with them?

We’ve invested around $30,000. 

What is the status of this engagement?

We began working with them in November 2019, and the engagement is ongoing.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement? 

We measured the success of the engagement by looking at the quality of the demo. We’re seeing success in terms of being able to make continuous improvements with each version we develop. 

How did IT-Jim perform from a project management standpoint?

Communication was fluid. We used Slack, email, and Zoom to communicate. We also worked in a cloud-based environment where we hosted our work so it was easy to share files. 

What did you find most impressive about them?

They’re experts in their field. We were able to take their advice and make gains based on that. They were also skilled in planning a research phase. 

Are there any areas they could improve? 

I don’t think there’s anything they could’ve done differently. I’m happy with what they delivered within our deadlines. 

5.0
Overall Score
  • 5.0 Scheduling
    ON TIME / DEADLINES
  • 5.0 Cost
    Value / within estimates
    They didn’t charge extra if the work needed to go over.
  • 5.0 Quality
    Service & deliverables
    They were able to solve the problem we came to them with.
  • 5.0 NPS
    Willing to refer

Custom Software Dev for SaaS Construction Platform

"Nothing we tried before came close to the quality and accuracy we’ve achieved with It-Jim."

Quality: 
5.0
Schedule: 
5.0
Cost: 
4.5
Willing to refer: 
5.0
The Project
 
$50,000 to $199,999
 
Jan. 2018 - Ongoing
Project summary: 

It-Jim built a symbol recognition engine to read blueprints for a construction estimation SaaS platform. The team worked alongside an existing engineering team and supplied expertise.

The Reviewer
 
11-50 Employees
 
Austin, Texas
Founder, SaaS Construction Platform
 
Verified
The Review
Feedback summary: 

The engine launched within six months, but the client continues to work with It-Jim on small tasks. It-Jim shows a high level of academic expertise and business sense in their work. The team makes collaboration easy with constant communication, flexible policies, and an openness to feedback.

A Clutch analyst personally interviewed this client over the phone. Below is an edited transcript.

BACKGROUND

Introduce your business and what you do there.

I’m the founder of a construction estimation SaaS platform. 

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with It-Jim?

We deal with a lot of blueprints that involve symbols and numbers, and we wanted to find a way to automatically count those symbols from the PDF image files. We tried to find a solution in-house and try some image recognition platforms, but we weren’t able to get the accuracy and results we needed. 

SOLUTION

What was the scope of their involvement?

It-Jim worked on the symbol recognition engine that underlies our products. They’re highly structured to work with. They gathered requirements and defined them. They delivered the desired output from a typical engineering sprint with great communication throughout. I think the team is primarily made of academics based on their work output and detailed artifacts. We appreciate that; it allows us to assess where we are in terms of future delivery.

What is the team composition?

Ievgen (CEO) is our main point of contact. He handled the scoping exercise and now interfaces with me and my engineering team. 

How did you come to work with It-Jim?

I found them by googling around for agencies that specialized in image recognition software. Our engineering team is in the Czech Republic, so I was looking for folks in Central and Eastern Europe to align with their time zones. It-Jim had a good portfolio, and Ievgen has a very academic background — he’s a member of the faculty at the University of Kharkiv, I believe — as well as business acumen.

How much have you invested with them?

We’ve spent about $75,000.

What is the status of this engagement?

We started working together at the beginning of 2018. It took six months for the initial delivery of the engine. We’ve subsequently done smaller projects with It-Jim on a one-off basis.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

One of the key goals we wanted to achieve was the engine’s accuracy: we wanted our customers to have confidence in the results we’re delivering. After a whole host of tests and tuning, we got to a very, very accurate engine. 

Working with It-Jim is very easy. There have been times where we pull in new deadlines or change the scope for certain reasons, and they are very accommodating. They make sure we achieve the right business outcomes without penalizing us financially.

How did It-Jim perform from a project management standpoint?

We use Slack and Skype to communicate, and we’ve given them access to our Jira. However, they do a self-managing on their side. Our communication was clear; we always understood when we were going to get something, and we got a heads up a few days in advance before something arrives. They've never missed a deadline.

What did you find most impressive about them?

I’ve worked with similar contractors and am used to them having a defined structure and attention of detail. But It-Jim takes that to the next level. Their technical ability defines them when it comes to symbol recognition and image manipulation. We tried other companies and industry SaaS platform, and nothing we tried before came close to the quality and accuracy we’ve achieved with It-Jim.

Are there any areas they could improve?

I don’t think so. Our original engagement was flawless, and so were our follow-up projects. 

Any advice for potential customers?

It-Jim is a detail-oriented group. Be very clear in your upfront requirements to accelerate delivery.

5.0
Overall Score
  • 5.0 Scheduling
    ON TIME / DEADLINES
  • 4.5 Cost
    Value / within estimates
  • 5.0 Quality
    Service & deliverables
  • 5.0 NPS
    Willing to refer

AI & Computer Vision Dev for Automotive Software Company

"I’ve stayed with them for one and a half years because of the success of the work."

Quality: 
5.0
Schedule: 
5.0
Cost: 
4.5
Willing to refer: 
5.0
The Project
 
Confidential
 
Dec. 2018 - Ongoing
Project summary: 

It-Jim provides ongoing development support in the fields of algorithms, AI and deep learning, and product strategy. They’ve become a true partner, growing alongside their client and filling necessary roles.

The Reviewer
 
51-200 Employees
 
Israel
David Lempert
R&D Manager, Automotive Software Company
 
Verified
The Review
Feedback summary: 

Operating with a large degree of autonomy, It-Jim consistently delivers on requirements and maintains top-level communication. They’ve developed a pattern of following through on deadlines, which has helped them earn the client’s trust and respect and increase their word-of-mouth reach.

A Clutch analyst personally interviewed this client over the phone. Below is an edited transcript.

BACKGROUND

Introduce your business and what you do there.

We work with OEMs and Tier-1 manufacturers from Israel, Europe, and the rest of the world. We develop algorithms, software, and hardware, including machine vision systems. I’m the R&D manager of the company.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with It-Jim?

We’ve worked with them on the algorithms and software side of our business.

SOLUTION

What was the scope of their involvement?

I have 6–7 algorithm developers on my team, and I go to It-Jim when I need extra resources. They work on new algorithms and improve or optimize the ones made by our internal team.

My company is divided into two parts: AI and deep learning, and classical computer vision. I’ve also worked with It-Jim on both of these topics. They’re experts in AI and deep learning. They know the field well and have a lot of knowledge in it.

Thirdly, I go to them when we need a new idea or way of thinking about something. After each project, they send us full documentation of what they did, what they could improve and the timelines for those, and how we can go forward.

What is the team composition?

We work with their project manager, their CTO, and their algorithm team. I don’t know whether the algorithm people are also the software people or whether they have separate roles. They have C++, Python, and MATLAB developers.

How did you come to work with It-Jim?

It was either through Upwork or through the recommendation from one of my friends in Israel.

What is the status of this engagement?

We started working with It-Jim around December 2018, and it’s ongoing.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

We make statements of work for all the projects we start, stating with our success criteria for them. It-Jim has always met the success criteria we wrote for them.

I’m managing seven teams, so I’m not always aware of the details. Every three months, I ask the people working with It-Jim how things are going, and they always tell me that everything is good. I don’t have a bad comment to give.

How did It-Jim perform from a project management standpoint?

They’ve visited us in Israel, and we have a very good vibe. They’re very good guys. We have good communication with the team members, as well as the CTO and the project manager.

They really stand by the timelines they give us, and they make the algorithm optimizations they promise. We use email, WhatsApp, Jira, Teams, Microsoft Project, and phone calls to communicate. Ievgen (CEO, It-Jim) is available 24/7 on WhatsApp or over the phone.

What did you find most impressive about them?

They communicate clearly and they respect deadlines. I’d worked with a few companies in Ukraine, Bulgaria, and Belgium before starting with It-Jim, and I’ve stayed with them for one and a half years because of the success of the work. I normally don’t take the time to give referrals when asked for them, but I’ve done it for It-Jim.

Are there any areas they could improve?

I don’t have something for them to improve. I want to believe that our companies will grow together. Israeli companies use software houses in Ukraine and Belgium, and my dream is that we’ll grow together with It-Jim and turn into a big company.

Do you have any advice for future clients of theirs?

With any provider, everything that needs to be done should be written down, in terms of success criteria, timeline, and price. That way, everyone on both sides will understand what needs to be done.

5.0
Overall Score
  • 5.0 Scheduling
    ON TIME / DEADLINES
  • 4.5 Cost
    Value / within estimates
  • 5.0 Quality
    Service & deliverables
  • 5.0 NPS
    Willing to refer