Deep Learning stack I am constantly asked about the role Deep Learning plays in sports performance tracking and analysis, so...
SoccerNet Camera Calibration Challenge 2023 Task This technical write-up describes some details of the winning submission to the SoccerNet Camera...
Testing The post continues the series of benchmark rusults published previously with results for GPUs of the latest four generations...
Multitask learning Multi-task artificial neural networks has several advantages over a set of single-task neural networks. Sharing of features between...
Testing The post presents results of Turing and Pascal GPUs benchmarking with a popular Deep Learning Benchmark. PyTorch based tests...
This post contains a description of the 14th place solution by Argus team (Ruslan Baikulov, Nikolay Falaleev) of the TGS...
The goal The overall goal of the competition on Kaggle platform was to build an algorithm that automatically and accurately...
Speedup tricks The basic prediction script was profiled with yappi multithreaded python profiler, and the profiling results gave rise to...
Software and Hardware The model used in the submission as well as previous experiments were prepared with use of different...
The goal The overall goal of the Challenge is to generate a binary labeled image masks for vehicles and a...
Postprocessing of predictions It was observed that the predicted class distribution in the Kaggle: IEEE’s Signal Processing Society - Camera...
About the task The goal of the competition on Kaggle platform was to build an algorithm that identifies which camera...
Introduction I was faced with the necessity of computational power increase in order to meet my needs in computer vision...
Introduction The main goal of the project is to train an fully convolutional neural network (encoder-decoder architecture with skip connections)...