DBN Models for Visual Tracking and Prediction

2007 ◽  
pp. 176-193
Author(s):  
Qian Diao ◽  
Jianye Lu ◽  
Wei Hu ◽  
Yimin Zhang ◽  
Gary Bradski
Keyword(s):  
Visual Tracking ◽  
General Framework ◽  
Broad Class ◽  
Probability Distributions ◽  
Prediction Method ◽  
Dynamic Bayesian Networks ◽  
Time Series Models ◽  
Complex Environments ◽  
Inference Algorithms ◽  
Background Clutter

In a visual tracking task, the object may exhibit rich dynamic behavior in complex environments that can corrupt target observations via background clutter and occlusion. Such dynamics and background induce nonlinear, nonGaussian and multimodal observation densities. These densities are difficult to model with traditional methods such as Kalman filter models (KFMs) due to their Gaussian assumptions. Dynamic Bayesian networks (DBNs) provide a more general framework in which to solve these problems. DBNs generalize KFMs by allowing arbitrary probability distributions, not just (unimodal) linear-Gaussian. Under the DBN umbrella, a broad class of learning and inference algorithms for time-series models can be used in visual tracking. Furthermore, DBNs provide a natural way to combine multiple vision cues. In this chapter, we describe some DBN models for tracking in nonlinear, nonGaussian and multimodal situations, and present a prediction method to assist feature extraction part by making a hypothesis for the new observations.

Effect of background clutter on neural discrimination in the bat auditory midbrain

2021 ◽  
Author(s):  
Kathryne M Allen ◽  
Angeles Salles ◽  
Sanwook Park ◽  
Mounya Elhilali ◽  
Cynthia F. Moss
Keyword(s):  
Acoustic Scattering ◽  
Object Discrimination ◽  
Complex Environments ◽  
Acoustic Features ◽  
Temporal Separation ◽  
Complex Sounds ◽  
Electrophysiological Recordings ◽  
Background Clutter ◽  
Spatio Temporal ◽  
Neural Underpinnings

The discrimination of complex sounds is a fundamental function of the auditory system. This operation must be robust in the presence of noise and acoustic clutter. Echolocating bats are auditory specialists that discriminate sonar objects in acoustically complex environments. Bats produce brief signals, interrupted by periods of silence, rendering echo snapshots of sonar objects. Sonar object discrimination requires that bats process spatially and temporally overlapping echoes to make split-second decisions. The mechanisms that enable this discrimination are not well understood, particularly in complex environments. We explored the neural underpinnings of sonar object discrimination in the presence of acoustic scattering caused by physical clutter. We performed electrophysiological recordings in the inferior colliculus of awake big brown bats, to broadcasts of pre-recorded echoes from physical objects. We acquired single unit responses to echoes and discovered a sub-population of IC neurons that encode acoustic features that can be used to discriminate between sonar objects. We further investigated the effects of environmental clutter on this population's encoding of acoustic features. We discovered that the effect of background clutter on sonar object discrimination is highly variable and depends on object properties and target-clutter spatio-temporal separation. In many conditions, clutter impaired discrimination of sonar objects. However, in some instances clutter enhanced acoustic features of echo returns, enabling higher levels of discrimination. This finding suggests that environmental clutter may augment acoustic cues used for sonar target discrimination and provides further evidence in a growing body of literature that noise is not universally detrimental to sensory encoding.

scholarly journals Identification problem for stochastic models with application to carcinogenesis, cancer detection and radiation biology

2002 ◽  
Vol 7 (3) ◽  
pp. 177-189 ◽  
Author(s):  
L. G. Hanin
Keyword(s):  
Parameter Identification ◽  
Tumor Progression ◽  
Cell Survival ◽  
Cancer Detection ◽  
Stochastic Models ◽  
General Framework ◽  
Broad Class ◽  
Identification Problem ◽  
Radiation Biology ◽  
Unified Approach

A general framework for solving identification problem for a broad class of deterministic and stochastic models is discussed. This methodology allows for a unified approach to studying identifiability of various stochastic models arising in biology and medicine including models of spontaneous and induced Carcinogenesis, tumor progression and detection, and randomized hit and target models of irradiated cell survival. A variety of known results on parameter identification for stochastic models is reviewed and several new results are presented with an emphasis on rigorous mathematical development.

scholarly journals Extraction of 3D Features from Complex Environments in Visual Tracking Applications

2007 ◽  
Author(s):  
Marta Marron ◽  
Juan Carlos Garcia ◽  
Miguel Angel Sotelo ◽  
Daniel Pizarro Perez ◽  
Ignacio Bravo Munoz
Keyword(s):  
Visual Tracking ◽  
Complex Environments

Effective template update mechanism in visual tracking with background clutter

2020 ◽  
Author(s):  
Shuai Liu ◽  
Dongye Liu ◽  
Khan Muhammad ◽  
Weiping Ding
Keyword(s):  
Visual Tracking ◽  
Template Update ◽  
Background Clutter

scholarly journals Adaptive Context-Aware and Structural Correlation Filter for Visual Tracking

2019 ◽  
Vol 9 (7) ◽  
pp. 1338 ◽  
Author(s):  
Bin Zhou ◽  
Tuo Wang
Keyword(s):  
Visual Tracking ◽  
High Performance ◽  
State Of The Art ◽  
Correlation Filter ◽  
Context Aware ◽  
Partial Occlusion ◽  
Structural Correlation ◽  
Background Clutter ◽  
Overall Performance ◽  
Fast Motion

Accurate visual tracking is a challenging issue in computer vision. Correlation filter (CF) based methods are sought in visual tracking based on their efficiency and high performance. Nonetheless, traditional CF-based trackers have insufficient context information, and easily drift in scenes of fast motion or background clutter. Moreover, CF-based trackers are sensitive to partial occlusion, which may reduce their overall performance and even lead to failure in tracking challenge. In this paper, we presented an adaptive context-aware (CA) and structural correlation filter for tracking. Firstly, we propose a novel context selecting strategy to obtain negative samples. Secondly, to gain robustness against partial occlusion, we construct a structural correlation filter by learning both the holistic and local models. Finally, we introduce an adaptive updating scheme by using a fluctuation parameter. Extensive comprehensive experiments on object tracking benchmark (OTB)-100 datasets demonstrate that our proposed tracker performs favorably against several state-of-the-art trackers.

scholarly journals HKSiamFC: Visual-Tracking Framework Using Prior Information Provided by Staple and Kalman Filter

Sensors ◽  
2020 ◽  
Vol 20 (7) ◽  
pp. 2137 ◽  
Author(s):  
Chenpu Li ◽  
Qianjian Xing ◽  
Zhenguo Ma
Keyword(s):  
Kalman Filter ◽  
Visual Tracking ◽  
Prior Information ◽  
State Of The Art ◽  
The Other ◽  
Robust Tracking ◽  
Similarity Learning ◽  
Complex Environments ◽  
Color Information ◽  
Learning Problem

In the field of visual tracking, trackers based on a convolutional neural network (CNN) have had significant achievements. The fully-convolutional Siamese (SiamFC) tracker is a typical representation of these CNN trackers and has attracted much attention. It models visual tracking as a similarity-learning problem. However, experiments showed that SiamFC was not so robust in some complex environments. This may be because the tracker lacked enough prior information about the target. Inspired by the key idea of a Staple tracker and Kalman filter, we constructed two more models to help compensate for SiamFC’s disadvantages. One model contained the target’s prior color information, and the other the target’s prior trajectory information. With these two models, we design a novel and robust tracking framework on the basis of SiamFC. We call it Histogram–Kalman SiamFC (HKSiamFC). We also evaluated HKSiamFC tracker’s performance on dataset of the online object tracking benchmark (OTB) and Temple Color (TC128), and it showed quite competitive performance when compared with the baseline tracker and several other state-of-the-art trackers.

Estimating Measurement Error Based on Evidence Theory

2012 ◽  
Vol 548 ◽  
pp. 839-842
Author(s):  
Ming Zhu Xiao
Keyword(s):  
Measurement Error ◽  
General Framework ◽  
Probability Distributions ◽  
Evidence Theory ◽  
Belief Functions ◽  
Random Set ◽  
Estimation Process

Measurement error is traditionally represented with probability distributions. Although probabilistic representations of measurement error have been successfully employed in many analyses, such representations have been criticized for requiring more refined knowledge with respect to the existing error than that is really present. As a result, this paper proposes a general framework and process for estimating the measurement error based on evidence theory. In this research cumulative belief functions (CBFs) and cumulative plausibility functions (CPFs) are used to estimate measurement error. The estimation includes two steps:(1) modeling the parameters by means of a random set, and discrediting the random set to focal elements in finite numbers; (2)summarizing the propagation error. An example is demonstrated the estimation process.

Fuzzy Detection aided Real-time and Robust Visual Tracking under Complex Environments

2020 ◽  
pp. 1-1 ◽  
Author(s):  
Shuai Liu ◽  
Shuai Wang ◽  
Xinyu Liu ◽  
Chin-Teng Lin ◽  
Zhihan Lv
Keyword(s):  
Real Time ◽  
Visual Tracking ◽  
Complex Environments

scholarly journals On random quadratic forms: supports of potential local maxima

2018 ◽  
Vol 55 (4) ◽  
pp. 1113-1130 ◽  
Author(s):  
Boris Pittel
Keyword(s):  
Population Genetics ◽  
Quadratic Forms ◽  
Hazard Rate ◽  
High Probability ◽  
Stable Equilibrium ◽  
Broad Class ◽  
Probability Distributions ◽  
Local Equilibrium ◽  
Local Maximum ◽  
Local Maxima

Abstract The selection model in population genetics is a dynamic system on the set of the probability distributions 𝒑=(p1,…,pn) of the alleles A1…,An, with pi(t+1) proportional to pi(t) multiplied by ∑jfi,jpj(t), and fi,j=fj,i interpreted as a fitness of the gene pair (Ai,Aj). It is known that 𝒑̂ is a locally stable equilibrium if and only if 𝒑̂ is a strict local maximum of the quadratic form 𝒑T𝒇𝒑. Usually, there are multiple local maxima and lim𝒑(t) depends on 𝒑(0). To address the question of a typical behavior of {𝒑(t)}, John Kingman considered the case when the fi,j are independent and [0,1]-uniform. He proved that with high probability (w.h.p.) no local maximum may have more than 2.49n1∕2 positive components, and reduced 2.49 to 2.14 for a nonbiological case of exponentials on [0,∞). We show that the constant 2.14 serves a broad class of smooth densities on [0,1] with the increasing hazard rate. As for a lower bound, we prove that w.h.p. for all k≤2n1∕3, there are many k-element subsets of [n] that pass a partial test to be a support of a local maximum. Still, it may well be that w.h.p. the actual supports are much smaller. In that direction, we prove that w.h.p. a support of a local maximum, which does not contain a support of a local equilibrium, is very unlikely to have size exceeding ⅔log2n and, for the uniform fitnesses, there are super-polynomially many potential supports free of local equilibriums of size close to ½log2n.

Sign in / Sign up

Export Citation Format

Share Document