English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The importance of human action recognition technology has been increasing1). Action recognition techniques based on video analysis have many applications, such as for motion-based video searches or video summarization. In particular, with the rapid spread of surveillance cameras, there is growing interest in developing a system that not only detects people but also automatically tracks and identifies their actions. However, it is very difficult to analyze real surveillance videos. Video taken by surveillance cameras is typically low quality and low resolution. Noisy background regions make it hard to perform motion analysis. Luminance changes and occlusions also interrupt the image regions that show people's actions. The diversity of human actions is another big issue. The speeds or joint movements for a certain action will differ greatly between individuals. Even the movements of the same person will strongly depend on the circumstance. For the above reasons, automatic recognition of such actions in a real video sequence has been viewed as a challenging task. In this paper, we discuss a method to robustly recognize specific human actions from video sequences taken in crowded places. Fig.1 shows sample images in the dataset of the TRECVID Surveillance Event Detection (SED) task2)3). TRECVID is an international workshop using real video corpora for evaluating information retrieval technologies. The task uses real surveillance videos that were shot in a crowded airport. The images contain many occlusions and variety Abstract There is growing interest in automatic recognition of human actions in video sequences shot by surveillance cameras. However, it's difficult to analyze human actions in real environments. That is, almost all of the current techniques can only detect simple actions within video sequences showing controlled environments. We propose action recognition methods based on multiple trajectories that can identify human actions within crowd sequences of real surveillance video. The methods use novel techniques for detecting diverse actions: a motion-speed invariant feature descriptor made from a key-point trajectory, and a weighting and clustering for the trajectory features. We conducted several experiments on the proposed methods, in which our previously proposed single-trajectory method was used as a baseline for comparison and the dataset was that of the TRECVID Surveillance Event Detection task. We discuss how to select the proper method to detect actions in crowd situations through an analysis of these experimental results.

[Paper] Using Trajectory Features to Recognize Human Actions Within Crowd Sequences of Real Surveillance Video