Computer Vision Still Retains its Appeal

Like many AI enthusiasts, my journey also began with a fascination for Computer Vision.
​
My journey into AI, started with my academic pursuit to obtain an MSc in Machine Learning, Computer Vision and Space Robotics.

The masters was tough, really tough. But inbetween long lectures tackling complex topics like calculus, kinematics, and traditional machine learning techniques seemed like an insurmountable mountain. But amidst the challenging academic hustle, something clicked—Deep Learning

For reasons still not entirely clear to me, the intersection of Computer Vision and the application of deep learning models to tasks such as face detection, pose estimation, and object detection, resonated deeply. It was as if all the puzzle pieces fell right into place, the patterns and logic making perfect sense.

Shortly after completing my MSc, I landed my first AI role as a Computer Vision Engineer. And as they say, there was no looking back. Within a few months, I found myself authoring numerous articles on computer vision techniques and deep learning models. Fast forward less than two years from graduation, and I was sharing my knowledge on the O'Reilly platform, teaching computer vision and deep learning to practitioners around the world.

So, here are my takeaways:

• The field of AI is vast. You neither can nor should aim to be an expert in every nook and cranny. Find a niche within AI that not only makes sense to you but also piques your curiosity. Dive deep into it, whether it's training pose estimation models in your spare time or devouring research papers.
• Learn in public. In this digital age, don't underestimate the power of social learning. Share your insights and newfound knowledge with the world using platforms like LinkedIn, Twitter, Substack, Medium, or YouTube.

Articles To Read

My newest article dives into the cutting-edge neural architecture that's the current buzz in the computer vision community: YOLO-NAS.

In this piece, I dissect the intricate technical aspects of the YOLO-NAS architecture, touching upon design considerations and training procedures that contributed to the development of this highly accurate and efficient object detection model.

Click on the link below to delve into the details, and feel free to share your thoughts. I'm always open to engaging in insightful discussions.

YOLO-NAS Uncovered: Essential Insights and Implementation Techniques for Machine Learning Engineers | by Richmond Alake | Jun, 2023 | Medium

Podcast

My latest podcast is a conversation I had with Eva Esteban, an Embedded Software Engineer at OpenBCI who is on the cutting edge of developing biosensors integrated with VR headsets. Eva shares her wealth of experience working on the development of firmware and software for Galea, a VR headset with biosensor integration that includes EEG, EMG, EDA, PPG, and eye-tracking.

​

​

​

Thanks for reading.