Tracking Objects in YouTube Videos
ثبت نشده
چکیده
Object tracking is an area of active research in Computer Vision. Tracking is a challenging problem, fundamentally due to the loss of information caused by the projection of a 3D scene on a 2D image. In this project, we are interested in tracking objects in YouTube videos, which are very diverse and have high variation in terms of changes in illumination, scale of objects, viewpoint, intra class variablity, etc. making it even more complicated. In this project, we briefly explore different techniques for object tracking and focus on one particular technique, "Tracking-LearningDetection" (TLD) [1]. An open source implementation of TLD, OpenTLD [2], was tested and some of its capabilities which are useful for tracking objects in YouTube videos have been noted, along with the changes needed to make it capable of being used effectively for this project. Some of these vital improvements were made to OpenTLD, like making it view point invariant, and the resulting performance has been compared to the original version.
منابع مشابه
Impact of Using YouTube Videos on Iranian Intermediate Male and Female EFL Learners' Listening Skill Development
This study aimed at investigating the effect of instructional YouTube videos on Iranian male and female EFL learners' listening comprehension. It was conducted with 60 male and female intermediate male and female EFL learners. The Oxford Quick Placement Test (OQPT) was used to measure the participants’ language proficiency to make sure about homogeneity of the participants. The participants wer...
متن کاملObject Detection, Tracking, and Motion Segmentation for Object-level Video Segmentation
We present an approach for object segmentation in videos that combines frame-level object detection with concepts from object tracking and motion segmentation. The approach extracts temporally consistent object tubes based on an off-the-shelf detector. Besides the class label for each tube, this provides a location prior that is independent of motion. For the final video segmentation, we combin...
متن کاملGrouping-Based Low-Rank Trajectory Completion and 3D Reconstruction
Extracting 3D shape of deforming objects in monocular videos, a task known as non-rigid structure-from-motion (NRSfM), has so far been studied only on synthetic datasets and controlled environments. Typically, the objects to reconstruct are pre-segmented, they exhibit limited rotations and occlusions, or full-length trajectories are assumed. In order to integrate NRSfM into current video analys...
متن کاملAre Health Videos from Hospitals, Health Organizations, and Active Users Available to Health Consumers? An Analysis of Diabetes Health Video Ranking in YouTube
Health consumers are increasingly using the Internet to search for health information. The existence of overloaded, inaccurate, obsolete, or simply incorrect health information available on the Internet is a serious obstacle for finding relevant and good-quality data that actually helps patients. Search engines of multimedia Internet platforms are thought to help users to find relevant informat...
متن کاملMonocular Multiview Object Tracking with 3D Aspect Parts
In this work, we focus on the problem of tracking objects under significant viewpoint variations, which poses a big challenge to traditional object tracking methods. We propose a novel method to track an object and estimate its continuous pose and part locations under severe viewpoint change. In order to handle the change in topological appearance introduced by viewpoint transformations, we rep...
متن کامل