Artificial Intelligence & Computer Vision

Work in artificial intelligence spans more than a decade, beginning during early university studies and evolving into long-term research and applied engineering practice. From early exploration of neural networks to modern deep learning systems, this work reflects sustained engagement with the field.

Experience includes applied machine learning and deep learning across computer vision and data-driven systems, covering model development, experimentation, and practical system design for real-world environments.

Today, this work is shaped by the intersection of academic research and industry application, with a focus on building robust, interpretable, and scalable AI systems that operate reliably within complex technical and organizational contexts.

Projects

Smart Assistant for Intelligent Colon Cancer Segmentation

Synthetic Data Generator and Vehicle Semantic Segmentation

Domains of interest

Semantic Segmentation
Instance Segmentation
Object Detection and Localization
Image Classification
Medical Imaging
Tumor Segmentation
Dataset Development
Transfer Learning

Tech stack & Tools

Python 3
Keras
Tensor Flow
OpenCV
NumPy
Pandas
Matplotlib
MITK Workbench
ImageJ
MicroDicom
MRIcron

EDU Materials

Besides everything that interests me and I am doing, my intentions are related to knowledge sharing, also. I am planning to create different EDU materials to help both younger colleagues and experienced engineers outside the DL domain also to come to grips with many different topics, from the prerequisites of Deep Learning such as Python programming, ML/DL and Computer Vision basics, databases, different useful libraries, to the Deep Learning architectures, data sets, training process and the tricks of the trade of Deep Learning and Computer Vision domain.

Useful Topics

How To Build A Successful AI PoC | Sicara

A 2021 guide to Semantic SegmentationIntroDeep learning has been very successful when working with images as data and is currently at a stage where it works better than humans on multiple use-cases. The most important problems that humans have been interested in solving with computer vision are image classification, object detection and segmentation in the

ML Optimization Methods and TechniquesFind out why ML optimization is necessary and how ML optimization techniques work through simple, practical examples.

TensorFlow 2.0 Tutorial : Optimizing Training Time Performance | SicaraTutorial to improve TensorFlow training time with tf.data pipeline optimizations, mixed precision training and multi-GPU strategies

Introducing tf-explain, Interpretability for TensorFlow 2.0 | SicaraUnderstanding deep networks is crucial for model development and user adoption. tf-explain offers interpretability methods to gain insight on your network.

Applying Occam’s razor to Deep Learning - KDnuggetsFinding a deep learning model to perform well is an exciting feat. But, might there be other -- less complex -- models that perform just as well for your application? A simple complexity measure based on the statistical physics concept of Cascading Periodic Spectral Ergodicity (cPSE) can help us be…

Set up TensorFlow with Docker + GPU in Minutes | SicaraThis tutorial aims to install a complete toolkit for real-time object recognition using deep learning.

Technology Readiness Levels DemystifiedTechnology readiness levels, or TRLs, give innovators a roadmap to follow as they try to take a technology from an idea to a proven benefit that's ready to go to work in the real world.

What are Technology Readiness Levels? | Mentor Works Ltd.TRLs are a nine-point scale that’s used to determine technology readiness. Use it to better guide research projects and tap into government funding programs.

Technology Readiness LevelTechnology Readiness Levels (TRL) are a type of measurement system used to assess the maturity level of a particular technology.

Using Technology Readiness Levels to Manage RiskIn a previous blog “Are the Requirements Attainable?”, we mentioned that one reason requirements may be not be attainable was because the ability to meet a given requirement was based on an immature technology. Engineers, scientists, and marketing personnel often … Read More

Image Segmentation in 2021: Architectures, Losses, Datasets, and Frameworks | Neptune BlogFind out what people do in image segmentation in 2020. Popular architectures, losses, datasets, frameworks and more. Everything you need to know about segmentation!

Find Open Datasets and Machine Learning Projects | KaggleDownload Open Datasets on 1000s of Projects + Share Projects on One Platform. Explore Popular Topics Like Government, Sports, Medicine, Fintech, Food, More. Flexible Data Ingestion.

Google Colab Vs Kaggle Kernels: Which Of The Two Platforms Should You Go For?Google has two free cloud platforms for GPUs — Google Colab and Kaggle Kernels. Here are the differences in specific features for the two.

Hyperparameter Tuning in Python: a Complete Guide 2021 - neptune.aiChoosing the correct hyperparameters for machine learning or deep learning models is one of the best ways to extract the last juice out of your models. In this article, I will show you some of the best ways to do hyperparameter tuning that are available today (in 2021). What is the difference between parameter and […]

Transfer Learning Guide: A Practical Tutorial With Examples for Images and Text in Keras - neptune.aiIt can take weeks to train a neural network on large datasets. Luckily, this time can be shortened thanks to model weights from pre-trained models – in other words, applying transfer learning. Transfer learning is a technique that works in image classification tasks and natural language processing tasks. In this article, you’ll dive into: what […]

Machine Learning Model Management: What It Is, Why You Should Care, and How to Implement It - neptune.aiMachine learning is on the rise. With that, new issues keep popping up, and ML developers along with tech companies keep building new tools to take care of these issues. If we look at ML in a very basic way, we can say that ML is conceptually software with a bit of added intelligence but […]

A Comprehensive Guide On How to Monitor Your Models in Production - neptune.ai*Record scratch* *Freeze frame* Yup, that’s me being plowed to the ground because the business just lost more than $500,000 with our fraud detection system by wrongly flagging fraudulent transactions as legitimate, and my boss’s career is probably over. And that guy in the chair? That’s our DevOps Engineer. You’re probably wondering how we got […]

Experiment Tracking vs Machine Learning Model Management vs MLOps - neptune.aiIt takes quite a lot of steps to take a machine learning model from idea to production. These steps can get too complex, too quickly. In this article, we’ll focus on dissecting the three main aspects of model deployment. These are: experiment tracking machine learning model management MLOps (machine learning operations) At the end of […]

Project Managing Readiness Before developing a project plan, a project manager must first understand the team's capability to implement the project, an understanding which requires knowing the team's delivery process and its areas of expertise. This paper examines a methodology for practicing project readiness. In doing so, it defines the concept of project readiness and the significance--to a project's outcome--of assembling and preparing a project team to perform their tasks and complete their responsibilities; it discusses the process of practicing project readiness, noting the activities that can help teams prepare for their project assignments. It describes three challenges in practicing project readiness; it lists the key project responsibilities performed by the project sponsor and the readiness manager. It identifies a readiness team's key players and a readiness assessment's primary criteria. It then outlines a readiness management model, detailing its major components in relation to the readiness life cycle and explaining the

Training Sets, Validation Sets, and Holdout SetsLearn about data partitioning with training, validation, and holdout and how it contributes to more accurate predictive models with automated machine learning.

Understanding memory usage in deep learning models training | SicaraShedding some light on the causes behind CUDA out of memory ERROR, and an example on how to reduce by 80% your memory footprint with a few lines of code in Pytorch

Minimizing real-time prediction serving latency in machine learning

Evaluating image segmentation models.When evaluating a standard machine learning model, we usually classify our predictions into four categories: true positives, false positives, true negatives, and false negatives. However, for the dense prediction task of image segmentation, it's not immediately clear what counts as a "true positive" and, more generally, how we

7 Tools to Build Proof-of-Concept Pipelines for Machine Learning Applications - neptune.aiIt always takes at least a few steps to convert any idea into reality. Sometimes, specialists can rely on wireframes or models to prove that an idea is relevant and potentially valuable. But depending only on wireframes is risky. That’s why we also do POC’s. As the name suggests, Proof of Concept is the minimal […]

Best Tools to Do ML Model Monitoring - neptune.aiWhy should you monitor your model? There are many reasons. It can help you understand the accuracy of your predictions, prevent prediction errors, and tweak your models to perfect them. Overall, ML model monitoring is necessary for making your model successful. One of the easiest ways to ensure things work smoothly is to use ML […]

Best Tools to Do ML Model Serving - neptune.aiTools for model serving in machine learning can provide you with solutions to many of the data engineers and devops concerns. They have many functionalities that make it easier to manage your models. You can use them during the entire lifecycle of your ML project, beginning with building a trained model, to deploying, monitoring, providing […]

How to Organize Deep Learning Projects - Examples of Best Practices - neptune.aiFor a successful deep learning project, you need a lot of iterations, a lot of time, and a lot of effort. To make this process less painful, you should try to use your resources to the max. A good step-by-step workflow will help you do that. With it, your projects become productive, reproducible, and understandable. […]

Best Machine Learning Model Management Tools That You Need to Know - neptune.aiDeveloping your model is an essential part of working on ML projects. And it’s usually a tough challenge. Every data scientist has to face it, along with difficulties, like losing track of experiments. These difficulties are likely to be both annoying and unobvious, which will make you feel confused from time to time. That’s why […]

Page updated

Google Sites

Report abuse