Using Local Steering Kernels to Detect People in Videos
نویسندگان
چکیده
Locally adaptive regression kernels (LARK) can be used as effective descriptors of generic objects, and have been used in object detection algorithms that do not require prior training [1]. This paper demonstrates how these kernels can be applied to the problem of people detection, taking into account the fact that a person can appear in a wide variety of poses, and can come in different shapes and sizes.
منابع مشابه
Object Recognition based on Local Steering Kernel and SVM
The proposed method is to recognize objects based on application of Local Steering Kernels (LSK) as Descriptors to the image patches. In order to represent the local properties of the images, patch is to be extracted where the variations occur in an image. To find the interest point, Wavelet based Salient Point detector is used. Local Steering Kernel is then applied to the resultant pixels, in ...
متن کاملFaceRipper Automatic Face Indexer and Tagger for Personal Albums and Videos
Digitized multimedia data do not have any structure in raw form and so is not amenable for information extraction by a machine. Consider the problem of creating structure over images by tagging it with labels which can easily be processed by machines and meaningful to humans. Being faces most important part of images, we consider problem of tagging them and indexing images or videos based on fa...
متن کاملA New Method for Sperm Detection in Human Semen: Combination of Hypothesis Testing and Local Mapping of Wavelet Sub-Bands
Introduction Automated methods for sperm characterization in microscopic videos have some limitations such as: low contrast of the video frames and possibility of neighboring sperms to touch each other. In this paper a new method is introduced for detection of sperms in microscopic videos. Materials and Methods In this work, first microscopic videos are captured from specimens of human semen. S...
متن کاملA New Method for Characterization of Biological Particles in Microscopic Videos: Hypothesis Testing Based on a Combination of Stochastic Modeling and Graph Theory
Introduction Studying motility of biological objects is an important parameter in many biomedical processes. Therefore, automated analyzing methods via microscopic videos are becoming an important step in recent researches. Materials and Methods In the proposed method of this article, a hypothesis testing function is defined to separate biological particles from artifact and noise in captured v...
متن کاملPeople Detection and Re-identification for Multi Surveillance Cameras
Re-identifying people in a network of non overlapping cameras requires people to be accurately detected and tracked in order to build a strong visual signature of people appearances. Traditional surveillance cameras do not provide high enough image resolution to iris recognition algorithms. State of the art face recognition can not be easily applied to surveillance videos as people need to be f...
متن کامل