📰 Medium · Deep Learning

1. Introduction: Replacing the Human Driver Continue reading on Medium »

By Day 5, I understood why WGAN uses a Critic instead of a traditional Discriminator. Continue reading on Medium »

Removing the background from an image used to mean opening an editor, tracing around the subject, and cleaning up the edges by hand. As of… Continue reading on

A practical journey from depth estimation and 3D reconstruction to semantic understanding Continue reading on Medium »

How AI learns to see, understand, and reason about the visual world—from Face ID to Multimodal AI. Continue reading on Medium »

From self-driving cars to real-time surveillance, object detection is at the core of modern computer vision — and YOLOv8 is leading that… Continue reading on Me

How a machine learned to see and why it thinks the way you do Continue reading on Medium »

When I first started learning Computer Vision, I assumed every vision model had the same objective: detect objects in an image. Continue reading on Medium »

Author: Saqib Ali Continue reading on Medium »

Every CV project eventually reaches a moment where someone has to decide: is this good enough to ship? Not “is it perfect” — nothing is… Continue reading on Med

Learn how to build an object detection system that locates and classifies dice using CNN-based feature extraction, bounding box regression… Continue reading on

Two families of methods currently dominate 3D reconstruction from images, and they are built on opposite assumptions. Continue reading on Medium »

This is the first blog in an upcoming series documenting the build of a computer vision project — from raw data to a complete webapp… Continue reading on Toward

Sebelum kita melanjutkan tutorial, ada baiknya kita mengetahui dulu istilah-istilah yang akan digunakan di step berikutnya. Pada postingan… Continue reading on

There’s a pattern that holds across almost every computer vision project I’ve seen or worked on: the cost of a mistake isn’t fixed. It… Continue reading on Medi

Picture a coach standing on the sideline, watching a match from one specific angle. Now picture a drone hovering directly overhead… Continue reading on Medium »

Somewhere along the way I noticed that the hardest part of a computer vision project usually wasn’t getting the model to work. It was… Continue reading on Mediu

Bir güvenlik kamerasının kalabalık içinde insanları anında tespit ettiğini, otonom bir aracın yayaları ve trafik levhalarını gerçek… Continue reading on Medium

Why I felt like cheating feeding raw 3D points into a neural net until I understood the math behind it. Continue reading on Medium »

I used to think the difference between teams that shipped CV products successfully and teams that stalled out was model quality. Better… Continue reading on Med

If you asked me a year ago how long a CV project takes, I’d have given you an answer based on how long the model takes to train. Continue reading on Medium »

If you asked me a year ago how long a CV project takes, I’d have given you an answer based on how long the model takes to train. Continue reading on Medium »

Why Vision Transformers quietly eat your tokens — told from the inside. Continue reading on Medium »

Abstract. The forward pass turns pixels into a prediction; this final part runs the same path backwards — how a single number, the loss… Continue reading on Med