KwanHua Lee

head-pose-estimation. Real-time head pose estimation built with OpenCV and dlib

openpose-darknet. Openpose implementation using darknet

v4l2-framebuffer. Map frame from USB camera to Linux framebuffer

mtcnn-head-detection. Head detection with modified cascaded CNN

dog-bark-detector. Detect dog bark from spectrogram classification using sndfile-tools and darknet

facenet-darknet-inference. Face recognition using facenet

face-detection-C. face detection using classical Viola-Jones method

sound-feature-extraction-C. Useful feature extraction for next step classification

darknet-clblast. Run darknet using OpenCL backend with CLBlast sgemm

spectrogram-opencv. Spectrogram display with the help of sndfile-tool, opencv, and portaudio

face-point-shape-model-tracking. ancient code with deprecated function, refactoring required

eye-isophote-curvature. paper: Accurate Eye Center Location and Tracking Using Isophote Curvature

darknet-inference-only. Simplfied Darknet inference steps

nnapi-sample-convolution. Run 2D convolution layer with NNAPI

mobilenetssd-C-scratch. zero CNN framework

NN-inference-edge. C++

FaceSwap. Real-time FaceSwap application built with OpenCV and dlib

pyAudioAnalysis. Python Audio Analysis Library: Feature Extraction, Classification, Segmentation and Applications

opencv-haarcascade-visualizer. visualize weak classifiers in each stage

active-contour. Using GVF (gradient vector flow) as external force

EasyOpenPose. OpenPose的简单实现