Pinned Repositories
analyze_ada_hessian
Analyze AdaHessian optimizer on 2D functions.
CTCDecoder
Connectionist Temporal Classification (CTC) decoding algorithms: best path, beam search, lexicon search, prefix search, and token passing. Implemented in Python.
CTCWordBeamSearch
Connectionist Temporal Classification (CTC) decoder with dictionary and language model.
DeslantImg
The deslanting algorithm sets text upright in images. Python, C++ and OpenCL implementations provided.
GPUImageProcessingArticle
HTRPipeline
Detect and read handwritten words on scanned pages.
SimpleHTR
Handwritten Text Recognition (HTR) system implemented with TensorFlow.
WordDetector
Detect handwritten words (classic image processing based method).
WordDetectorNN
Detect handwritten words (neural network based).
githubharald's Repositories
githubharald/SimpleHTR
Handwritten Text Recognition (HTR) system implemented with TensorFlow.
githubharald/CTCDecoder
Connectionist Temporal Classification (CTC) decoding algorithms: best path, beam search, lexicon search, prefix search, and token passing. Implemented in Python.
githubharald/CTCWordBeamSearch
Connectionist Temporal Classification (CTC) decoder with dictionary and language model.
githubharald/WordDetector
Detect handwritten words (classic image processing based method).
githubharald/DeslantImg
The deslanting algorithm sets text upright in images. Python, C++ and OpenCL implementations provided.
githubharald/HTRPipeline
Detect and read handwritten words on scanned pages.
githubharald/WordDetectorNN
Detect handwritten words (neural network based).
githubharald/analyze_ada_hessian
Analyze AdaHessian optimizer on 2D functions.
githubharald/GPUImageProcessingArticle
githubharald/cpp_mnist
A neural network written in ~260 lines of plain C++ code without any dependencies.
githubharald/fast_inv_sqrt
Fast approximation of the inverse square root 1/√(x).
githubharald/ParticleSystem
A simple particle system implemented with Python and OpenCL.
githubharald/quasi_cauchy_optimizer
Implementation of the quasi Cauchy optimizer, an optimization method from the quasi Newton family. It uses a diagonal approximation of the Hessian and therefore has a small memory footprint.
githubharald/githubharald.github.io
githubharald/separate_kernel
Separate a 2D convolution kernel into two 1D kernels.
githubharald/analyze_pooling