Doug Fenstermacher's personal website
Most stopwatches assume one runner and one clock. Epochron is built for coaches and timers tracking multiple people at once. It supports group stopwatches for relays, heats, or ranked comparisons, and get live split predictions mid-race instead of doing the math yourself. It runs entirely in the browser, works offline (built for track meets with no signal), and keeps all data on your device. No…
A deeper dive into modern approaches for generating aesthetically pleasing lines based on difference-of-Gaussians (DoG) edge detections.
Sharing an attempt at computational hatching. Using structure tensors, edge tangent flow, and streamline integration to transform photographs into hand-drawn-looking illustrations.
Learn how to implement effective prompt analytics for large language models while safeguarding user privacy. Discover key strategies, best practices, and ethical considerations in this guide.
Discover how Neural Variational Document Models, implemented using PyTorch, improve topic modeling and unsupervised learning in natural language processing. Learn about the architecture, training process, and applications of these latent variable models for text analysis and beyond.
Discover how to harness the power of the X-Means clustering algorithm and CIE2000 color distance metric to accurately extract dominant colors from images. This advanced technique combines unsupervised machine learning with human color perception principles to generate visually appealing and representative color palettes. Perfect for data-driven design, image analysis, and computer vision…
Exploring complexities of optimizing web performance by implementing K-Means clustering algorithms using WebAssembly and Emscripten.
Improving meet performance lists in track & field using the previous performances of the competitors
A Typescript library for physical fitness calculations
Simplifying polylines and trajectories using the Ramer-Douglas-Puecker algorithm, and summarizing graphs/networks to speed up computations using grouping-based and compression-based algorithms.