Christoph Heindl's Projects
Low-cost 3D scanner using Laser Triangulation
A PyTorch-Lightning callback to increase model reproducibility through enforcing consistent git repository states upon training and validation.
Poisson image editing for seamless cloning and other operations
Experiments involving projective geometry of circles and how that affects computer vision tasks
Code for "Enhanced Human-Machine Interaction by Combining Proximity Sensing with Global Perception" IROS 2019
A pure PyTorch based implementation of "Instant Neural Graphics Primitives with a Multiresolution Hash Encoding" with tweaks.
:tangerine: Intro to symbolic computation in Python including applications to function optimization, physics simulation and more. Includes notebooks on back-propagation, auto-diff and more.
Various classic artificial intelligence algorithms applied to common problems.
PID controller playground in Python
Dimensional analysis and modeling in Python
Gaussian processes samples
Python implementation of "Global Data Association for MOT Tracking using Network Flows"
Ridid body physics simulation in Python
Utilities for infrared signal debugging
Fast linear assignment problem (LAP) solvers for Python based on c-extensions
Fast simulation of mass-spring systems
Exemplary implementations of statistical methods for sampling from probability distributions
A modern Python library to work with Anoto dot patterns.
Minimal, standalone, fast Python OpenEXR reader for single-part, uncompressed scan-line files as produced by Blender.
:bar_chart: Benchmark multiple object trackers (MOT) in Python
Comparing a couple of integrators for solving ODEs
Probabilistic robot localization
Blueprint code for projects combining flask/aiohttp and AngularJS in a single web application
:art: Artistic neural style transfer with tweaks in PyTorch
Code for computing interpolating / approximating thin plate splines.
Capture tone sequences from real-time audio streams in Python
Detect and extract time overlays in videos.
Seamlessly integrate numpy arrays into pydantic models.
Minimal HLS streaming demo with dynamic marker support in Python
:sweat_drops: Seamless, distributed, real-time integration of Blender into PyTorch data pipelines