Loading contributions...
Top Repositories
Reconstructs a 3D stripe on the area of an object on which a laser falls as seen by the camera
Greenscreen Matting done on still images and videos
Various camera models (Full, Weak and Orthographic) are used to convert 3D real world points into 2D image pixel coordinates by simulating a 'virtual camera'
A shot is a continious sequence of frames filmed in one go. The regions of discrepency between two shots is called the shot boundary. Various discriminatory features like the frame variance, mean, L2-norm etc are taken (as a means of representing video frame data) and then various classifiers are used to compare their performances in detecting these shot boundaries.
Does camera calibration, rectification, correspondence and finally 3D reconstruction
Contains code for poisson compositing and it's modification to do inpainting by second order PDE's
Repositories
38Reconstructs a 3D stripe on the area of an object on which a laser falls as seen by the camera
Various camera models (Full, Weak and Orthographic) are used to convert 3D real world points into 2D image pixel coordinates by simulating a 'virtual camera'
Run a simulation to make robots run around a bit world
Greenscreen Matting done on still images and videos
Implements Pyramidal blending wherein a laplacian pyramid is created out of portions of images to be blended and then the pyramids themselves added per level to give a new third pyramid, which is then superimposed to give a new image
Contains code for poisson compositing and it's modification to do inpainting by second order PDE's
Small GUI demo to see how changing various parameters of the hough circle transform (written in python and OpenCV)
Does camera calibration, rectification, correspondence and finally 3D reconstruction
Detection and Tracking for an AR marker
A shot is a continious sequence of frames filmed in one go. The regions of discrepency between two shots is called the shot boundary. Various discriminatory features like the frame variance, mean, L2-norm etc are taken (as a means of representing video frame data) and then various classifiers are used to compare their performances in detecting these shot boundaries.
Does triangulation matting (created for the CVFX Final Project, Fall 2017 RPI)
No description provided.
The one and only! The tale! The legend! Tomasi and Kanade's original factorization algorithm which gave rise to 3D reconstruction! Amazing!
Large scale pattern mining
Simple number guessing game which advertises its instance using Avahi / Zeroconf.
Simple setup for a motion detection and tracking system using background subtraction and kalman filtering
Finds all english synonyms in conceptNet for the Machine Teaching Project @CISL RPI
Finds the path between any two given concepts in ConceptNet (under construction) for the Machine Teaching project @CISL RPI
This is a simple TFTP server. It is based on the RFC 1350. This was developed and tested on Ubuntu 16 and 14.04.
Utterance Parser written for the Machine Teaching project @CISL RPI
Alternate means of analysing line plots, done for the Narrative Generation project @CISL RPI
Does camera calibration using point to point correspondences
Implementation for the Independent Multimodal Background Subtraction based on a paper written by Bloisi and Iocchi
My solutions to the hacker-rank code challenges.
Practising tensorflow using examples found online
Made for the Intelligent Analysis Room, CISL, RPI
Analyzing dialogue acts from meeting data in various ways
A complete digital prototype for the Harmony Mills Part of the AR Mural Interaction, made for the City of Cohoes. Starts with the user scanning the marker to gain access to actual webVR experience which takes them to an interactive panorama
Aframe relies on AR.js to make WebVR applications. I'm uploading my own attempt at learning to use Aframe or at least the basics of it
Determines optical flow values on a per voxel basis utilizing cubic facet fitting to approximate the intensities of the image