Results 1 - 10
of
14
Action Recognition from One Example
, 2009
"... We present a novel action recognition method based on space-time locally adaptive regression kernels and the matrix cosine similarity measure. The proposed method uses a single example of an action to find similar matches. It does not require prior knowledge about actions; foreground/background segm ..."
Abstract
-
Cited by 29 (1 self)
- Add to MetaCart
We present a novel action recognition method based on space-time locally adaptive regression kernels and the matrix cosine similarity measure. The proposed method uses a single example of an action to find similar matches. It does not require prior knowledge about actions; foreground/background segmentation, or any motion estimation or tracking. Our method is based on the computation of novel space-time descriptors from a query video, which measure the likeness of a voxel to its surroundings. Salient features are extracted from said descriptors and compared against analogous features from the target video. This comparison is done using a matrix generalization of the cosine similarity measure. The algorithm yields a scalar resemblance volume, with each voxel indicating the likelihood of similarity between the query video and all cubes in the target video. Using nonparametric significance tests and non-maxima suppression, we detect the presence and location of actions similar to the query video. High performance is demonstrated on challenging sets of action data containing fast motions, varied contexts, and even when multiple complex actions occur simultaneously within the field of view. Further experiments on the Weizmann and KTH datasets demonstrate state-of-the-art performance in action categorization, despite the use of only a single example.
Volumetric features for video event detection
, 2008
"... Real-world actions occur often in crowded, dynamic environments. This poses a difficult challenge for current approaches to video event detection because it is difficult to segment the actor from the background due to distracting motion from other objects in the scene. We propose a technique for eve ..."
Abstract
-
Cited by 21 (0 self)
- Add to MetaCart
(Show Context)
Real-world actions occur often in crowded, dynamic environments. This poses a difficult challenge for current approaches to video event detection because it is difficult to segment the actor from the background due to distracting motion from other objects in the scene. We propose a technique for event recognition in crowded videos that reliably identifies actions in the presence of partial occlusion and background clutter. Our approach is based on three key ideas: (1) we efficiently match the volumetric representation of an event against oversegmented spatio-temporal video volumes; (2) we augment our shape-based features using flow; (3) rather than treating an event template as an atomic entity, we separately match by parts (both in space and time), enabling robustness against occlusions and actor variability. Our experiments on human actions, such as picking up a dropped object or waving in a crowd show reliable detection with few false positives. 1.
An attention based method for motion detection and estimation
- Workshop on Computational Attention and Applications, ICVS
, 2007
"... The demand for automated motion detection and object tracking systems has promoted considerable research activity in the field of computer vision. A novel approach to motion detection and estimation based on visual attention is proposed in the paper. Comparisons are made with the Berkeley MPEG-1 vid ..."
Abstract
-
Cited by 2 (2 self)
- Add to MetaCart
(Show Context)
The demand for automated motion detection and object tracking systems has promoted considerable research activity in the field of computer vision. A novel approach to motion detection and estimation based on visual attention is proposed in the paper. Comparisons are made with the Berkeley MPEG-1 video analyzer [1]. Preliminary results show that the new method extracts more information about the moving object than that available in the MPEG encoding. In addition the method does not suffer from some of the inaccuracies inherent in the MPEG encoding.
Motion detection using a model of visual attention
- in Proc. of ICIP
"... ABSTRACT Motion detection and estimation are known to be important in many automated surveillance systems. It has drawn significant research interest in the field of computer vision. This paper proposes a novel approach to motion detection and estimation based on visual attention. The method uses t ..."
Abstract
-
Cited by 2 (2 self)
- Add to MetaCart
(Show Context)
ABSTRACT Motion detection and estimation are known to be important in many automated surveillance systems. It has drawn significant research interest in the field of computer vision. This paper proposes a novel approach to motion detection and estimation based on visual attention. The method uses two different thresholding techniques and comparisons are made with Black's motion estimation technique [1] based on the measure of overall derived tracking angle. The method is illustrated on various video data on and results show that the new method can extract motion information. .
6 Fast detecting and tracking of moving objects in video scenes
, 2006
"... ha l-0 ..."
(Show Context)
2D Motion Description and Contextual Motion Analysis: Issues and New Models
, 2004
"... In this paper, several important issues related to visual motion analysis are addressed with a focus on the type of motion information to be estimated and the way contextual information is expressed and exploited. Assumptions (i.e., data models) must be formulated to relate the observed image intens ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
In this paper, several important issues related to visual motion analysis are addressed with a focus on the type of motion information to be estimated and the way contextual information is expressed and exploited. Assumptions (i.e., data models) must be formulated to relate the observed image intensities with motion, and other constraints (i.e., motion models) must be added to solve problems like motion segmentation, optical flow computation, or motion recognition. The motion models are supposed to capture known, expected or learned properties of the motion field, and this implies to somehow introduce spatial coherence or more generally contextual information. The latter can be formalized in a probabilistic way with local conditional densities as in Markov models. It can also rely on predefined spatial supports (e.g., blocks or pre-segmented regions). The classic mathematical expressions associated with the visual motion information are of two types. Some are continuous variables to represent velocity vectors or parametric motion models. The other are discrete variables or symbolic labels to code motion detection output (binary labels) or motion segmentation output (numbers of the motion regions or layers). We introduce new models, called mixed-state auto-models, whose variables belong to a domain formed by the union of discrete and continuous values, and which include local spatial contextual information. We describe how such...
A SALIENCY BASED OBJECT TRACKING METHOD
"... A novel three-stage framework for object tracking under stationary background conditions is proposed in this paper. The first stage uses an attention based method to extract motion information. The second stage then applies a region growing and matching technique to motion vectors to obtain motion s ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
A novel three-stage framework for object tracking under stationary background conditions is proposed in this paper. The first stage uses an attention based method to extract motion information. The second stage then applies a region growing and matching technique to motion vectors to obtain motion segmentation. Finally the moving objects are tracked based on the displacements of region centroids. The method is tested on various real-world video data and empirical results show that the proposed approach can track moving objects and extract motion information from non-rigid objects such as moving people without prior knowledge of the object’s size or shape. 1.
MULTISCALE NEIGHBORHOOD-WISE DECISION FUSION FOR REDUNDANCY DETECTION IN IMAGE PAIRS
, 2010
"... Abstract. To develop better image change detection algorithms, new models able to capture spatio-temporal regularities and geometries present in an image pair are needed. In this paper, we propose a multiscale formulation for modeling semi-local inter-image interactions and detecting local or region ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
Abstract. To develop better image change detection algorithms, new models able to capture spatio-temporal regularities and geometries present in an image pair are needed. In this paper, we propose a multiscale formulation for modeling semi-local inter-image interactions and detecting local or regional changes in an image pair. By introducing dissimilarity measures to compare patches and binary local decisions, we design collaborative decision rules that use the total number of detections obtained from the neighboring pixels, for different patch sizes. We study the statistical properties of the non-parametric detection approach that guarantees small probabilities of false alarms. Experimental results on several applications demonstrate that the detection algorithm (with no optical flow computation) performs well at detecting occlusions and meaningful changes for a variety of illumination conditions and signal-to-noise ratios. The number of control parameters of the algorithm is small and the adjustment is intuitive in most cases.
