jacobchen1998 Goto Github PK

cdc-press-release-crawler

deep-feature-encoding-with-mnist-data

UMAP is used to present the difference between the original MNIST data and its encoded features.

emoji-cheat-sheet

A markdown version emoji cheat sheet

feature-tracking-with-pca

Traditional feature tracking techniques such as SIFT, SURF, and Lucas Kanade algorithms define key points in terms of finding poles and cannot specify specific tracking points. The general Deep Learning based tracking algorithms such as Siamese tracker require a lot of resources for neural network training. Here, we implement feature tracking using PCA. Our algorithm can specify tracking points and does not require extensive training.

human-falling-detect-tracks

AlphaPose + ST-GCN + SORT.

involution_example_mnist

A simple example that uses involution layer instead of convolution layer in MNIST classification task.

iou-tracker

Python implementation of the IOU Tracker

muscle-enhancing-via-autoencoder

This is just a funny project that we want to see AutoEncoder (AE) can actually work to enhance the features we want. We will start to improve...

pixel-scanning-removal-hands11k-dataset

python-execute-matlab-code

A simple tutorial which teaches you how to call MATLAB in python.

real-time-yolov8-pose-tracking-sequence

scan-based-feature-tracking-c-

This is the new repository that is same concept as my previous project " Feature-tracking-with-PCA " but be written in C++

self-adaptive-feature-tracking-with-pca-lucas-kanade

The inability to change size has always been a drawback of sliding window tracking. If the previous frame of the current frame is used as the reference frame, the error rate is often superimposed. If only traditional feature tracking methods such as SIFT, SURF or Lucas-Kanade are used, it is not possible to track a specific object and there is no defined object frame to define the overall features of the object to be tracked. Using Deep Learning (DL) for object tracking such as Siamese Tracker requires training of the object to be tracked, and the size of the tracking bounding box cannot be defined arbitrarily while tracking. We propose to use Principal Component Analysis (PCA) as the feature extraction mechanism and Lucas-Kanade (LK) tracking optical flow as the object size prediction: 1. no time-consuming DL training is required for the objects. 2. 2. The object frame size can be defined arbitrarily. 3. 3. Automatically detects and adjusts object size even if it changes.

streaming-with-django-and-opencv

videoaddsubtitles_with_whisper

End to end add (translated) subtitles on video file with whisper which developed by OpenAI.