Recognition of Marker-less Human Actions in Videos Using Hidden Markov Models

نویسندگان

  • David O. Johnson
  • Arvin Agah
چکیده

In this paper, we present a novel methodology for using Hidden Markov Models (HMM) to recognize marker-less human actions in videos. We use Scale Invariant Feature Transform (SIFT) keypoints to identify and track the trajectories of objects and major body parts in instructional videos. Then, we use a HMM to recognize the trajectories as a human action. This enables the use of readily available instructional videos from the Internet to train robots to perform tasks instead of programming them. The HMM we designed is able to recognize a marker-less human grasping an object in a video with a precision of 80%. The experiments also showed that the HMM can predict whether the human is reaching, is grasping, or has grasped the object with an accuracy of 75%.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Improving Phoneme Sequence Recognition using Phoneme Duration Information in DNN-HSMM

Improving phoneme recognition has attracted the attention of many researchers due to its applications in various fields of speech processing. Recent research achievements show that using deep neural network (DNN) in speech recognition systems significantly improves the performance of these systems. There are two phases in DNN-based phoneme recognition systems including training and testing. Mos...

متن کامل

MAN-MACHINE INTERACTION SYSTEM FOR SUBJECT INDEPENDENT SIGN LANGUAGE RECOGNITION USING FUZZY HIDDEN MARKOV MODEL

Sign language recognition has spawned more and more interest in human–computer interaction society. The major challenge that SLR recognition faces now is developing methods that will scale well with increasing vocabulary size with a limited set of training data for the signer independent application. The automatic SLR based on hidden Markov models (HMMs) is very sensitive to gesture's shape inf...

متن کامل

Belief Scheduler for sequential data analysis based on model failure detection in TBM framework applied to activity recognition in athletics videos

In this paper, a generic architecture based on the Transferable Belief Model (TBM) is proposed for on line and automatic human action and activity recognition in videos of athletics sports with moving camera. The first contribution of this paper is a tool called Belief Scheduler which makes belief on actions smoother, separates actions states (true or false) and synchronizes actions in order to...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2011