Ashish Pandey's starred repositories
Full-Body-Estimation-using-Media-Pipe-Holistic
A notebook showcasing how to use mediapipe holistic for body tracking.
MultiPoseMovenetLightning
A walkthrough demonstrating multi person tracking using movenet lighting
Geospatial-Analysis-of-Traffic-Navigation
How about adding some info that can quanitfy the stress on a particular navigation route using satellite feed and can help in toll operations & traffic management The current analysis is on the satellite image frame, but if we can add this info to a live feed it can help in developing smoother navigation systems based on real time traffic scenario
Social-Distancing-Detector
Computer vision has opened up a lot of opportunities to explore into AI domain that were earlier highly limited. Here is an application of haarcascade classifier and OpenCV to develop a social distancing violation detector. I am passing the algo through a video feed where it first detects people using 'haarcascade_fullbody.xml' classifier algo. OpenCV and some mathematical operations then allow us to make code the social distancing violation logic
Multiperson-Pose-Detection
Application of Movenet model to detect multiple persons in a frame using convolutional neural network model
Nanda-Devi-Glacier-Loss
Feburary 7,2021 Ecological Disaster (Nanda Devi Glacier, IND: 7,108 m above sea level). Satellite image analysis using the methodology of image segmentation shows that the Glacier cover in Nanda Devi has substantially decreased over the last 4 decades. It has gone down from 43% in Year 1984 to 20% in Year 2022 (in relation to the captured area in image)
Blob-Detection
This is an example of application of morphological operations on an image using the concepts of 1) creating an image to Brinary 2) Erosion and Dilution operations to bring out the key features in an image 3) Blob detection on virus image
Blob-Detection--Coronavirus-Envelope-Protien
Here is the application of Computer vision- Blob Detection methodology to detect and count the Envelope protiens of sample images of Human Coronavirus and mouse hepatitis virus (Image source- https://www.utmb.edu/virusimages/VI/coronaviruses and cdc)
Image-Segmentation--Impact-of-pace-of-Urbanisation-and-land-use
Application of Image Segmentation algorithms to understand “How sustainable is the pace of Urbanisation & Net land usage?” scoped on two geographical locations.
Multiple-Image-Classification--ResNet50
ResNet-50 is a convolutional neural network that is 50 layers deep. You can load a pretrained version of the network trained on more than a million images from the ImageNet database [1]. The pretrained network can classify images into 1000 object categories, such as keyboard, mouse, pencil, and many animals. As a result, the network has learned rich feature representations for a wide range of images. The network has an image input size of 224-by-224
Multiple-Image-classification--VGG-19
Classify multiple images in a folder with minimal code using keras API for VGG 19 image classification model.
Pose-Detection
Application of MediaPipe Holistic pipeline that integrates integrates separate models for pose, face and hand components
Pose-Estimation--Real-time-clap-count
MediaPipe Human Pose estimation to count the number of claps
Tutorial-Deep-Learning_Keras-Sequential
This is an application of Keras Sequential model on the Churn Modeling data set. Objective was to develop an end to end model to train the model+ also the steps to preprocess the test data to align it in the same shape as that of the train data set. The entire model can be used as a learning step to develop an deep Learning ML model using all the good stuffs from Scikit Learn (Pipelines, Feature extraction and Hyper parameter tuning).