C

computer vision

Robochan: middleware for interfacing between generic algorithms (controllers) and robotic platforms and environments (parrot, gym, robosim etc.)

Active contour (level-set) image processing engine written in modern C++.

Fluvel is a region-based active contour engine focused on clarity, performance, and clean architecture. The core processing logic is independent from the UI, allowing reuse as a standalone library.

The project is currently under active development.

A custom convolutional neural network (CNN) was developed from scratch and trained to recognize flower types using pcituires from the Flowers Recognition Dataset. The model’s architecture and hyperparameters were optimized to maximize recognition accuracy.

The custom CNN is compared against MobileNetV2, a pre-trained model fine-tuned using transfer learning. The comparison highlights the trade-offs between custom-built models and transfer learning approaches in terms of accuracy, training efficiency, and computational cost.

Pictures blurring API for a Panoramax instance.

Was previously hosted at https://github.com/cquest/sgblur

Practical tasks on Deep Learning (DL) and Neural Networks (NN). 🤖

The main project for OctoMY™ - Ready-to-run robot software

Uses inceptionV3 to detect if an image file is NSFW for JVM/Android. Supports .png, .jpg, .jpeg, .gif, and more.

Non-local pixel vlaue analyzing and edge detecting algorithm. Againsting edge detection in adversial envrionments like CATPCHAs with near 1/6 C performance.

idtracker.ai tracks up to 100 unmarked animals from videos recorded in laboratory conditions using artificial intelligence. Free and open source.

This project leverages the Brain Tumor Image database to develop a semantic segmentation model for detecting brain tumors. The model is based on an encoder-decoder U-Net architecture, which classifies each pixel in a medical image as either tumor or healthy tissue. The resulting segmentation is visualized as a binary mask, clearly delineating the tumor region from the surrounding healthy areas.

The model successfully localizes the tumor’s global position with high accuracy but the precise shape and boundaries remain approximate (moderate Dice coefficient). Extending the training phase with additional epochs could refine the segmentation quality, leading to a higher Dice coefficient and more accurate tumor delineation.

ICL : Intellisrc Common Library : This library can help you to write less code and speedup the development of your next great application. It is particularly designed to accomplish most of the common tasks required by backend services.

The VirtuaLearn3D++ open-source scientific software for 3D point clouds.

A computer vision solution implemented in Python that utilizes deep learning techniques to classify food imagery for dietary analysis and nutritional tracking.

Solving 4chan captcha

Parqly is a cutting-edge solution for automated access control and seamless payment processing in parking zones. The project bridges the gap between hardware and software by combining Computer Vision for license plate recognition with a robust web-based platform for instant user payments.

An AI-driven pipeline for detecting and quantifying cancer and inflammatory tissue on biopsy slides to support standardized diagnostics. This independent fork focuses on advanced ensembling (SMP <-> nnU-Net), probability fusion, and Test-Time Augmentation (TTA), providing a highly reproducible training and inference workflow.

WASSfast is the next-generation stereo processing pipeline for sea waves 3D reconstruction.

This repo will have all resources, labs, data which I use/d on Kaggle Network