Is it possible to create a random 2- dimensional shape using mathematical equations Or in software like 3D-max and AutoCAD? like this one:
Yes, it is definitely possible to create a random 2-dimensional shape using mathematical equations or software like 3D Max and AutoCAD.
Using Mathematical Equations:
Using 2D Software (e.g., 3D Max and AutoCAD):
In both cases, the randomness can be controlled to various extents, allowing you to fine-tune the level of randomness or repeatability of the generated shapes.
Keep in mind that while the shapes may appear random, they are still generated by deterministic algorithms or equations. For truly unpredictable shapes, you might want to explore generative adversarial networks (GANs) or other advanced machine learning techniques, but that goes beyond the scope of traditional mathematical equations and standard 2D software.
Dear Mr. Shabani,
Why not when there are more and more avenues available?
just as Mr. Omid stated using mathematical equations.
Yes, it is feasible to create a random two-dimensional shape using specialized tools. If you are interested in finding a method to generate a similar shape randomly, I would recommend exploring innovative models like Stable Diffusion XL, which has been released recently. Using platforms like Google Colab, this model can be customized to craft a similar image, based on a clearly defined prompt. If the idea is to design a deep learning model, there is a variety of alternatives that can be explored, including but not limited to Generative Adversarial Networks (GAN), Variational Autoencoders, and other methods grounded in deep learning.
From my personal experience, I designed a model that operates in two stages and is based on GANs. In the initial stage, I approximate random numbers with the help of the Wasserstein metric, to approximate the distribution of random data (origin) to the distribution of peripheral blood cells, thus producing a low-resolution image. The second stage applies the generative concept of GAN in an image-to-image approach. This approach enables me to bring the data distribution closer between one image and another, thereby improving its quality. To achieve this, I make use of the Pix2Pix technique, which allows a precise conversion of the characteristics of one image into another, maintaining some attributes while altering others. If you are intrigued to know more about this project, I invite you to read the publication where I detail the work carried out: Article Automatic generation of artificial images of leukocytes and ...