Pinned Repositories
teachablemachine-community
Example code snippets and machine learning code for Teachable Machine
video-retalking
[SIGGRAPH Asia 2022] VideoReTalking: Audio-based Lip Synchronization for Talking Head Video Editing In the Wild
stm32ai-modelzoo
AI Model Zoo for STM32 devices
3D_Reconstruction_twoView_PlaneSweep_DenseRecon
implemented two-view stereo and multi-view stereo algorithms to convert multiple 2D viewpoints into a 3D reconstruction of the scene.
Face_Mask_Detection_CNN
The model designed achieves an overall of 95% test and validation accuracy, then Haar-Cascade Face Detection method is employed to crop faces in each image and make classifications upon cropped images using the trained CNN model.
generative-ai-for-beginners
12 Lessons, Get Started Building with Generative AI 🔗 https://microsoft.github.io/generative-ai-for-beginners/
Multiple-Object-Tracking-Under-Noise-and-Occlusion
Employed augmented belief updates and model the positions and uncertainties of occluders and tracked targets using Gaussian distributions, subsequently applying Kalman Filters and Extended Kalman Filters for tracking. For occluders, we use a hand recognition algorithm to distinguish between static and dynamic occlusions (such as hands).
onrobot
OnRobot packages (https://wiki.ros.org/onrobot)
PRML
PRML algorithms implemented in Python
TemplateMatchWithGUI
Qt for GUI TemplateMatch
UpennGitttt's Repositories
UpennGitttt/Multiple-Object-Tracking-Under-Noise-and-Occlusion
Employed augmented belief updates and model the positions and uncertainties of occluders and tracked targets using Gaussian distributions, subsequently applying Kalman Filters and Extended Kalman Filters for tracking. For occluders, we use a hand recognition algorithm to distinguish between static and dynamic occlusions (such as hands).
UpennGitttt/TemplateMatchWithGUI
Qt for GUI TemplateMatch
UpennGitttt/3D_Reconstruction_twoView_PlaneSweep_DenseRecon
implemented two-view stereo and multi-view stereo algorithms to convert multiple 2D viewpoints into a 3D reconstruction of the scene.
UpennGitttt/Face_Mask_Detection_CNN
The model designed achieves an overall of 95% test and validation accuracy, then Haar-Cascade Face Detection method is employed to crop faces in each image and make classifications upon cropped images using the trained CNN model.
UpennGitttt/generative-ai-for-beginners
12 Lessons, Get Started Building with Generative AI 🔗 https://microsoft.github.io/generative-ai-for-beginners/
UpennGitttt/onrobot
OnRobot packages (https://wiki.ros.org/onrobot)
UpennGitttt/PRML
PRML algorithms implemented in Python