From Pixels to Blocks: A 3D Transformation System

The first release is published, go try it! Now it’s time to work on a Bird’s Eye View algorithm!

Hello, I finally managed to get a good approximation of the vanishing point to estimate depth using perspective. However, I’m not sure whether I should continue with this approach (which is quite tricky) or try a different one, perhaps a physics-informed neural network based on perceptual physics, or an AI model like Anything V2.
I hope you’re doing well. Have a great day of coding!

Hi,
We’ve started going deeper into image depth analysis. I implemented some algorithms to detect lines in the picture and determine which ones are parallel to each other.
Now, let’s try to detect parallelograms in the image to create a depth approximation based on perspective. Talk to you soon!

Hey, to get started, I simply reorganized the folder structure of my project by creating the classes ImageProcessing.hpp and Function.hpp. I think it will now be easier to develop our image processing logic. Have a great day!

Hi! This is a small devlog. I updated my README and fixed some issues with the camera and orbital trajectory. Now I’m going to focus on data and image processing, trying to find patterns that appear in many images and may indicate depth. Maybe I’ll create a depth-based calculation using these patterns and eventually transform a side-view image into a zenithal (top-down) view to convert landscapes into 3D maps. I hope you wrote some great code today too! Bye ;)

Hey, finally we have the Viridis gradient to represent the height (I will add colors like Magma and Plasma later).
Now the mission is to create a few algorithms to detect the most important features and try to reduce the resolution as much as possible.If you have any suggestions, I’ll gladly take them!
Thanks for reading, and have a nice day!!

Guess which painting is represented in the picture…

We did it! After a few issues with my vertex shader and glVertexAttribPointer, I finally managed to trace the edges of my cube (next step: without the middle line ;). Now I will implement a feature that modifies the colors based on height, and then go back to our image processing system.
I hope you’re enjoying your day. Have a great day of coding!!

Hey! I started implementing image processing features. Actually, I began with the simplest one: grayscale average. However, I ran into a big problem, the rendering isn’t very useful without shades and color variation :(

So, let’s dive into shaders and try to add some lighting!
Have a great day coding!

Hey guys, what’s up?

Quick update on my project: I’ve added a few features to my Shunting Yard algorithm, like parentheses and functions. I’ll soon update my RPN calculator to support functions so I can finally start working on the image analysis.

Have a great coding day!

Hello, I finally managed to create a user-input-based function visualizer. First, I parsed the user input using the Shunting Yard algorithm (it doesn’t accept functions like exp, sin, or cos for now). Next, I calculate the position of each cube for every frame using an RPN calculator.

You can admire a beautiful x^3 / y function.

I hope you’re doing well. Bye bye!

New day, new goal! I’ve started implementing function visualization. Currently, I’m working on the Shunting Yard algorithm to convert user input into an expression my code can understand. I hope you’re doing well—have a great day of coding!!

Big improvement! Now we can move through our 3D space using ZSQD and rotate using the arrow keys. Now, let’s start implementing the beginning of our algorithm to convert images into volumetric shapes. Catch you next time!

Hi guys, today I managed to create and visualize 3D shapes and a cube. I had a few issues with my vertex shader (the part that renders our vertices) and with the translation system using GLM. But now we’re starting to see our grid. It’s really exciting. Have a great day of coding!!

I finally made my square, and I tried to position it inside a double nested loop to place a square at a given position relative to the coordinates passed into the function. It was quite hard to do, but I learned a few new principles in OpenGL, and that’s really exciting.

The next step will be to create a grid of these squares and maybe convert them into 3D cubes. I hope you’ll enjoy what’s coming next…

Thanks for reading!

After an hour of configuration, I finally installed C++ with all packages like OpenGL ( thanks to vcpkg). Finally, a managed to create my first windows. The next step will be to create my cube based on triangles in a 3-dimensional coordinate system. I hope you will like it !!