Bio

Mark Seliaev is an undergraduate student graduating in 2020 with a major in business informatics at ITMO University, Saint Petersburg, Russia. After graduating from high school, he became obsessed with mathematics and programming, and started his journey toward data science. Currently, he is interested in computer vision and autonomous cars. Mark plans to pursue a master’s degree in statistics/machine learning, so the Wolfram Summer School will definitely help him acquire useful skills for his career.

Computational Essay

Image Classification: Basic Approaches »

Project: Smart Image Auto-cropping

Goal

The goal of this project is to find the optimal region of an image to crop based on its content.

Main Results in Detail

There were three approaches covered for recognizing the most important part of an image: neural network sensitivity maps, image keypoints and real-time object detection (YOLO).

Future Work

Future work may consist of building a neural network from scratch using public datasets (CAT2000, MIT300) and trying to get a high score.