ImageProcessing

Build a simple, reusable foundation for your video processing apps and alleviate the common challenges of video manipulation.

TensorFlow is one of the leading tools for training deep learning models. Outside that space, it may seem intimidating and unnecessary, but it has many creative uses—like producing highly effective adversarial input for black-box AI systems.

Machine learning, combined with some standard image processing techniques, can result in powerful video analysis tools.

In this article, Toptal Freelance Software Engineer Michael Karchevsky walks through a solution for a machine learning competition that identifies the species and lengths of any fish present in a given video segment.

Image processing algorithms are often very resource intensive due to fact that they process pixels on an image one at a time and often requires multiple passes. Successive Mean Quantization Transform (SMQT) is one such resource intensive algorithm that can process images taken in low-light conditions and reveal details from dark regions of the image.

In this article, Toptal engineer Daniel Angel Munoz Trejo gives us some insight into how the SMQT algorithm works and walks us through a clever optimization technique to make the algorithm a viable option for handheld devices.