Programming Languages: Python, C#

Machine Learning Frameworks: TensorFlow, PyTorch, scikit-learn, Keras

Data Processing: Pandas, NumPy

Version Control and Collaboration Systems: Git, Bitbucket, Confluence, Jira

Tools: PyCharm, Visual Studio, VS Code, Jupyter Notebook, Google Colab

I studied and compared algorithms for detecting and describing features on images, such as SIFT, SURF, BRIEF, ORB. Analyzed their effectiveness on distorted images and experimented with their implementations.

Researched color theory and algorithms for selecting RGB color values, in particular, the issue of color matching.

Studied different image hashing techniques (perceptual hashing, DCT, Wavelet hashing) to effectively detect duplicate or similar images.

Deep dived into the concept and implementation of image keypoints to detect and map features between images.

Studied and applied techniques such as Tf-idf, Word2vec, Transformers (BERT) and architectures related to LSTM, CRF and Attention.

I took Stanford's CS224n course, which included theory lectures and labs that provided me with a deep understanding of word processing.

Researched image comparison and text processing techniques such as pooling and transformers.

Implemented machine learning algorithms and models from scratch in Python and NumPy, which helped me better understand the inner workings of neural networks.

I have been actively using materials from Stanford open courses, watching video lectures and studying articles on topics like SIFT, CNN, BERT, which greatly deepened my understanding of complex topics in the field of computer vision and NLP.