Hybrid Super Vector with Improved Dense Trajectories for Action Recognition
نویسندگان
چکیده
With recent improved dense trajectory features (HOG, warped HOF, and warped MBH), we employ two advanced super vector methods, namely Fisher Vector (FV) and soft Vector of Locally Aggregated Descriptors (VLAD-K) to encode them separately. The two individual super vectors are concatenated into a Hybrid Super Vector, and a linear SVM classifier is used to predict labels. We achieve 87.46%1 in average accuracy of the three training/testing splits on the UCF101 dataset.
منابع مشابه
Combined Ordered and Improved Trajectories for Large Scale Human Action Recognition
Recently, a video representation based on dense trajectories has been shown to outperform other human action recognition methods on several benchmark datasets. The trajectories capture the motion characteristics of different objects, for example human bodies, in spatial and temporal dimensions. In dense trajectories, points are sampled at uniform intervals in space and time and then tracked usi...
متن کاملVideo Covariance Matrix Logarithm for Human Action Recognition in Videos
In this paper, we propose a new local spatiotemporal descriptor for videos and we propose a new approach for action recognition in videos based on the introduced descriptor. The new descriptor is called the Video Covariance Matrix Logarithm (VCML). The VCML descriptor is based on a covariance matrix representation, and it models relationships between different low-level features, such as intens...
متن کاملHuman Action Recognition Using Improved Salient Dense Trajectories
Human action recognition in videos is a topic of active research in computer vision. Dense trajectory (DT) features were shown to be efficient for representing videos in state-of-the-art approaches. In this paper, we present a more effective approach of video representation using improved salient dense trajectories: first, detecting the motion salient region and extracting the dense trajectorie...
متن کاملLearning features from Improved Dense Trajectories using deep convolutional networks for Human Activity Recognition
In this work, we tackle the problem of recognizing human activities by exploring methods for incorporating the state of the art improved dense trajectories into a deep learning framework. Specifically, we explore efficacy of several models trained using the action tubes sampled from dense trajectory. We performed experiments two different architectures, the first one that resembles bag of words...
متن کاملFusion of Dense SURF Triangulation Features and Dense Trajectory based Features
In this paper, we describe our method used to achieve our results which was submitted to the Recognition Task of the challenge. As for video features, we combined our proposed feature [1] and the dense trajectories based feature presented in [2]. We employed Fisher Vector encoding to represent videos using these features and trained multiclass linear SVMs to perform action recognition. We condu...
متن کامل