High-performance multiple object tracking based on YOLO, Deep SORT, and KLT 🚀

Problem Set solutions for the "Introduction to Computer Vision (ud810)" MOOC from Udacity

Vehicle speed estimation using the optical flow algorithm from a mono camera(CCTV)

Taking a Deeper Look at the Inverse Compositional Algorithm (CVPR 2019, Oral)

A generic pipeline to align multimodal image pairs from different sensors by extending Lucas-Kanade on feature maps. CVPR2021

Robust Local Optical Flow (RLOF)

Optical Flow estimation in pure Python

Determining optical flow by using Horn-Schunck method and Lucas-Kanade method

An implementation of several tracking algorithms based on Lucas Kanade algorithms

An implementation of optical flow using both the Lucas Kanade method as well as Horn Schunck.

C++/CUDA solutions, slides, and resources for Udacity's Intro to Computer Vision (UD810) course.

C++ implementation of Lucas-Kanade-Image-Alignment

Robert Barron optical flow code: slightest modifications for modern computers

Implementation of Lucas-Kanade tracker algorithm to track a moving car, face of a baby and running Usain Bolt

This repo includes solutions to all the 'in the class quizzes' and 7 problem sets of the Introduction to Computer Vision course (G Tech CS6476 - on Udacity)

Optical Flow and Deep Learning Use Cases

Consist of four different approaches for generating optical flow and can be demonstrated in Colab.

Implementation for our CVPR 2021 oral paper "PointNetLK Revisited".

In this repository, we deal with the task of video frame interpolation with estimated optical flow. To estimate the optical flow we use Lucas-Kanade algorithm, Multiscale Lucas-Kanade algorithm (with iterative tuning), and Discrete Horn-Schunk algorithm. We explore the interpolation performance on Spheres dataset and Corridor dataset.