Nonrigid Motion Analysis: Articulated and Elastic Motion

Aggarwal, J. K.; Cai, Qin; Liao, Wen-Hung; Sabata, Bikash

doi:10.1006/cviu.1997.0620

Cited by 105 publications

(45 citation statements)

References 95 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Starting from larger blobs, we take every object and search for the ideal layer in the list to assign this object to, if an allocation is made, the chosen layer is removed from the list. Because the Euclidean distance metric's allows dimensions with larger scales and variances to dominate the feature space, the cost function for allocating moving objects to their corresponding layers is based on the Mahalanobis metric measure using equation (1). The use of the Mahalanobis metric alleviates most of the Euclidean metric limitations, as it accounts automatically for the scaling of coordinate axes in the feature space.…”

Section: Moving Regions Correspondencementioning

confidence: 99%

See 1 more Smart Citation

People detection and recognition using gait for automated visual surveillance

Bouchrika

Nixon

2006

IET Conference on Crime and Security

View full text Add to dashboard Cite

In this paper, a computer vision system for automated visual surveillance in an unconstrained outdoor environment is described. We propose a method for tracking multiple moving objects based on shape-based feature correspondence between consecutive frames. We have explored a new approach for walking people detection and recognition based on their gait motion. The novelty of our approach is motivated by the latest research for people identification using gait. The gait signature is derived using a model-based method. The experimental results confirmed the robustness of our method to discriminate between single walking people, groups of people and vehicles with a detection rate of %100. Furthermore, the system is able to recognize walking people with a CCR of %92.

show abstract

Section: Moving Regions Correspondencementioning

confidence: 99%

“…Two main approaches are being used for gait recognition [1,14]: model-based and non-model based methods. For the first one, a priori shape model is established to match real images to this predefined model, and thereby extracting the corresponding features once the best match is obtained.…”

Section: Introductionmentioning

confidence: 99%

People detection and recognition using gait for automated visual surveillance

Bouchrika

Nixon

2006

IET Conference on Crime and Security

View full text Add to dashboard Cite

show abstract

“…In the model, V1 neurons estimate local speed and direction in five-frame, 256x256 pixel image sequences using spatiotemporal filters (e.g., [44]) 1 . Their direction selectivity is restricted to 12 distinct, Gaussian-shaped tuning curves.…”

Section: The Feedforward Motion Pyramidmentioning

confidence: 99%

“…Many recent reviews of various aspects of motion understanding have not made any mention of attentive processing of any kind [1,2,3,4,5]. The review by Aggarwal and Cai [6] includes one example of work that uses motion cues to segment an object and to affix an attentional window on to it.…”

Section: Introductionmentioning

confidence: 99%

Attending to Motion: Localizing and Classifying Motion Patterns in Image Sequences

Tsotsos

Pomplun

Liu

et al. 2002

Biologically Motivated Computer Vision

View full text Add to dashboard Cite

Abstract. The Selective Tuning Model is a proposal for modelling visual attention in primates and humans. Although supported by significant biological evidence, it is not without its weaknesses. The main one addressed by this paper is that the levels of representation on which it was previously demonstrated (spatial Gaussian pyramids) were not biologically plausible. The motion domain was chosen because enough is known about motion processing to enable a reasonable attempt at defining the feedforward pyramid. The effort is unique because it seems that no past model presents a motion hierarchy plus attention to motion. We propose a neurally-inspired model of the primate visual motion system attempting to explain how a hierarchical feedforward network consisting of layers representing cortical areas V1, MT, MST, and 7a detects and classifies different kinds of motion patterns. The STM model is then integrated into this hierarchy demonstrating that successfully attending to motion patterns, results in localization and labelling of those patterns.

show abstract

“…Moreover, human motion analysis has many challenging issues, because the highly flexible structure and self-occlusion of the human body mandates complicated processes for motion segmentation and analysis. The motion of the human body is known as a form of non-rigid and articulated motion [2]. Human gait motion is one of the most universal and complex of all human activities [3], and it can be decomposed into periodic components within a gait cycle.…”

Section: Introductionmentioning

confidence: 99%