CWYAlpha

Just another WordPress.com site

Thought this was cool: Max-Margin Early Event Detectors – CVPR 2012 Best Student Paper

leave a comment »


看到tombone博客上的消息,今年CVPR的优秀学生论文是牛津vgg实验室的 Professor Andrew Zisserman教授的学生Hoai, Minh的一篇文章。

Hoai, Minh是CMU的PhD,到vgg做博后,作品和运气都不错。vgg也真是牛啊,在cvpr上获奖,也驰骋eccv。

==

文章要解决的一个问题是如何提前检测一下视频中要发生的事情。当然,预测事情发生是个很难很难的事情,所以paper和实验就规约到一些基本的简单问题上来,例如人脸的表情变化、人体的动作变化。

这样,人脸,人体基本都是有结构的模板性的目标,要描述细节上的变化,当然要捕捉一些细节的结构化的信息,自然地,作者用到了Structural SVMs。Structural Learning是今年cvpr很热的一个话题,当然也是很有用的一个话题,经典的工作要数Cornell的Thorsten Joachims教授的作品了。

具体的细节看论文吧。

想法很好,方法很简单。

==

Max-Margin Early Event Detectors.
Hoai, Minh & De la Torre, Fernando
CVPR 2012

Abstract:
The need for early detection of temporal events from sequential data arises in a wide spectrum of applications ranging from human-robot interaction to video security. While temporal event detection has been extensively studied, early detection is a relatively unexplored problem. This paper proposes a maximum-margin framework for training temporal event detectors to recognize partial events, enabling early detection. Our method is based on Structured Output SVM, but extends it to accommodate sequential data. Experiments on datasets of varying complexity, for detecting facial expressions, hand gestures, and human activities, demonstrate the benefits of our approach. To the best of our knowledge, this is the first paper in the literature of computer vision that proposes a learning formulation for early event detection.

Early Event Detector Project Page (code available on website)

Overview

Simulating the sequential arrival of training data

We simulate the sequential arrival of training data and use partial events as positive training examples. We train a single event detector to recognize all partial events, but our method does more than augmenting the set of training examples.

Monotonicity requirement

Monotonicity requirement—the detection score of a partial event cannot exceed the score of an encompassing partial event. MMED provides a principled mechanism to achieve this monotonicity, which cannot be assured by a naive solution that simply augments the set of training examples.

Results

Detecting disgust Detecting Disgust. See video
Detecting fear Detecting Fear. See video

From left to right: the onset frame, the frame at which MMED fires, the frame at which SOSVM fires, and the peak frame.

Code

Download mmed-release-0.1 and follow instruction in ./src/README.html.

您可能也喜欢:


CVPR 2012 papers


CVPR2011录取结果


Conference Paper Awards—CVPR2010


CVPR 2011 Best Paper-优秀论文


Tomasz:cvpr 2011: highlights from day 1

无觅

相关文章

from 丕子: http://www.zhizhihu.com/html/y2012/3840.html

Written by cwyalpha

六月 23, 2012 在 4:53 上午

发表在 Uncategorized

发表评论

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / 更改 )

Twitter picture

You are commenting using your Twitter account. Log Out / 更改 )

Facebook photo

You are commenting using your Facebook account. Log Out / 更改 )

Google+ photo

You are commenting using your Google+ account. Log Out / 更改 )

Connecting to %s

%d 博主赞过: