Improved natural language processing designed for the Kabyle language, incorporating new learning models and tokenization algorithms.

A multi-modal classification project using image and audio modality. Implements one shot learning with Siamese network to determine if the given image-audio pair belong to the same class.

Exploring Machine Learning and Deep Learning techniques for healthcare applications through Jupyter notebooks

Quote-Detective is an improved version of a tool for automatic quotation detection in text leveraging natural language processing techniques.

A Quadrotor simulator for training neural networks to control quadrotor flights using a deep q learning algorithm.

Open-source project for semantic segmentation with SLAM, originated as a graduation work, now maintained by woodminus.

A compilation and brief summary of knowledge points in the field of artificial intelligence (including computer vision, natural language processing, recommendation systems, etc.) gleaned from readi…

A more efficient way of image segmentation using Minimum Spanning Forests

Hokuyo laser rangefinder for object detection on a surface for Unity, favouring multiple sensor use, featuring rect and radius detection modes.

[CVPR 2022] Advanced Amodal Segmentation Applying Out-of-Task and Out-of-Distribution Generalization with an Improved Bayesian Model

This PyTorch implementation helps to understand Neural Ordinary Differential Scenes (a research paper published in ECCV 2020). Hone your PyTorch skills by exploring and contributing to the project,…

A comprehensive benchmarking of various attention mechanisms used in Vision Transformers

Exploration and modeling of language with focus on CMI and Perplexity obtained from Unigram, Bigram and Trigram simulation on Codemix data

A project integrating YOLOv3 for object detection with a Microsoft HoloLens front-end. Encompasses two main repository concepts, YOLOv3 and HoloLens, Unity, Azure Cognitive Services.

Repository containing notebooks used for the coursework of 'Applied Deep Learning for Natural Language Processing' at TUM.