Building 3D event logs for video investigation

In scene investigation, creating a video log captured using a handheld camera is more convenient and more complete than taking photos and notes. By introducing video analysis and computer vision techniques, it is possible to build a spatio-temporal representation of the investigation. Such a representation gives a better overview than a set of photos and makes an investigation more accessible. We develop such methods and present an interface for navigating the result. The processing includes (i) segmenting a log into events using novel structure and motion features making the log easier to access in the time dimension, and (ii) mapping video frames to a 3D model of the scene so the log can be navigated in space. Our results show that, using our proposed features, we can recognize more than 70 percent of all frames correctly, and more importantly find all the events. From there we provide a method to semi-interactively map those events to a 3D model of the scene. With this we can map more than 80 percent of the events. The result is a 3D event log that captures the investigation and supports applications such as revisiting the scene, examining the investigation itself, or hypothesis testing.

Title: 

Building 3D event logs for video investigation
Authors: Dang, Trung Kien
Worring, Marcel
Bui, The Duy
Keywords: Video analysis
Story navigation
Scene investigation
3D model
Issue Date: 2015
Publisher: Kluwer Academic Publishers
Citation: Scopus
Abstract: In scene investigation, creating a video log captured using a handheld camera is more convenient and more complete than taking photos and notes. By introducing video analysis and computer vision techniques, it is possible to build a spatio-temporal representation of the investigation. Such a representation gives a better overview than a set of photos and makes an investigation more accessible. We develop such methods and present an interface for navigating the result. The processing includes (i) segmenting a log into events using novel structure and motion features making the log easier to access in the time dimension, and (ii) mapping video frames to a 3D model of the scene so the log can be navigated in space. Our results show that, using our proposed features, we can recognize more than 70 percent of all frames correctly, and more importantly find all the events. From there we provide a method to semi-interactively map those events to a 3D model of the scene. With this we can map more than 80 percent of the events. The result is a 3D event log that captures the investigation and supports applications such as revisiting the scene, examining the investigation itself, or hypothesis testing.
Description: Multimedia Tools and Applications, Volume 74, Issue 13, 2 June 2015, Pages 4617-4639
URI: http://link.springer.com/article/10.1007%2Fs11042-013-1826-9
http://repository.vnu.edu.vn/handle/VNU_123/32331
ISSN: 13807501
Appears in Collections:Bài báo của ĐHQGHN trong Scopus

Nhận xét

Đăng nhận xét

Bài đăng phổ biến từ blog này

Development and regulation of pedicel abscission in tomato

Hình ảnh
To shed unfertilized flowers or ripe fruits, many plant species develop a pedicel abscission zone (AZ), a specialized tissue that develops between the organ and the main body of the plant. Regulation of pedicel abscission is an important agricultural concern because pre-harvest abscission can reduce yields of fruit or grain crops, such as apples, rice, wheat, etc. Tomato has been studied as a model system for abscission, as tomato plants develop a distinct AZ at the midpoint of the pedicel and several tomato mutants, such as jointless, have pedicels that lack an AZ. This mini-review focuses on recent advances in research on the mechanisms regulating tomato pedicel abscission. Molecular genetic studies revealed that three MADS-box transcription factors interactively play a central role in pedicel AZ development. Transcriptome analyses identified activities involved in abscission and also found novel transcription factors that may regulate AZ activities. Another study ide...
Đọc thêm

Multi-class minimax probability machine

This paper investigates the multi-class Minimax Probability Machine (MPM). MPM constructs a binary classifier that provides a worst-case bound on the probability of misclassification of future data points, based on reliable estimates of means and covariance matrices of the classes from the training data points. We propose a method to adapt MPM to multi-class datasets using the one-against-all strategy. And then we introduce an optimal kernel for MPM for each specific dataset found by Genetic Algorithms (GA) [1]. The proposed method was evaluated on stomach cancer data. The obtained results are better and more stable than for using a single kernel. Title:  Multi-class minimax probability machine Authors:  Dang, Tat-Dat; Nguyen, Ha-Nam Keywords:  Genetic algorithms; Minimax probability machine; One-against-all; One-against-one Issue Date:  2009 Publisher:  H:Đại học Quốc gia Hà Nội Abstract:  This paper investigates the multi-class Minim...
Đọc thêm

Quản lý đội ngũ công chức, viên chức của Tổng cục Dự trữ Nhà nước

Hình ảnh
Quản lý CCVC là vấn đề hệ trọng, phức tạp và nhạy cảm vì đây là một nguồn nhân lực đặc biệt của quốc gia, là đội ngũ nhân danh quyền lực nhà nước trong hoạt động công vụ. Quản lý hiệu quả đội ngũ CCVC có ý nghĩa quan trọng để phát huy nội lực của nền hành chính nhà nước. Có thể nói, công tác quản lý CCVC ở nƣớc ta còn nhiều hạn chế, cần những ý kiến đóng góp để nâng cao chất luợng công tác quản lý CCVC. Vì vậy, quản lý CCVC đang là vấn đề đƣợc sự quan tâm của nhiều nhà nghiên cứu, nhà quản lý, đặc biệt là trong giai đoạn đẩy mạnh cải cách hành chính nhà nước như hiện nay. Title:  Quản lý đội ngũ công chức, viên chức của Tổng cục Dự trữ Nhà nước Other Titles:  Management of civil servants and employees in General Department of State Reserves Authors:  Vũ, Trường Giang Keywords:  Nguồn nhân lực Quản lý Công chức Viên chức Issue Date:  2016 Publisher:  Đại học quốc gia Hà Nội Citation:  105 tr. URI:  http://repository.vnu.edu.vn...
Đọc thêm