Adaptive Fusion Of Particle Filtering And Spatio-Temporal Motion Energy For Human Tracking

Object tracking is an active research area nowadays due to its importance in human computer interface, teleconferencing and video surveillance. However, reliable tracking of objects in the presence of occlusions, pose and illumination changes is still a challenging topic. In this paper, we introduce a novel tracking approach that fuses two cues namely colour and spatio-temporal motion energy within a particle filter based framework. We conduct a measure of coherent motion over two image frames, which reveals the spatio-temporal dynamics of the target. At the same time, the importance of both colour and motion energy cues is determined in the stage of reliability evaluation. This determination helps maintain the performance of the tracking system against abrupt appearance changes. Experimental results demonstrate that the proposed method outperforms the other state of the art techniques in the used test datasets.

Clinical Characteristics of Reticular Pseudodrusen in the Fellow Eye of Patients with Unilateral Neovascular Age-Related Macular Degeneration

Purpose: To describe associations between reticular pseudodrusen, individual characteristics, and retinal function.

Design: Cohort study.

Participants: We recruited 105 patients (age range, 52–93 years) who had advanced neovascular age-related macular degeneration (AMD) in only 1 eye from 3 clinical centers in Europe.

Methods: Minimum follow-up was 12 months. The eye selected for study was the fellow eye without advanced disease. Clinical measures of vision were distance visual acuity, near visual acuity, and results of the Smith-Kettlewell low-luminance acuity test (SKILL). Fundus imaging included color photography, red-free imaging, blue autofluorescence imaging, fluorescein angiography, indocyanine green angiography, and optical coherence tomography using standardized protocols. These were used to detect progression to neovascular AMD in the study eye during follow-up. All imaging outputs were graded for the presence or absence of reticular pseudodrusen (RPD) using a multimodal approach. Choroidal thickness was measured at the foveal center and at 2 other equidistant locations from the fovea (1500 μm) nasally and temporally. Metrics on retinal thickness and volume were obtained from the manufacturer-supplied automated segmentation readouts.

Main Outcome Measures: Presence of RPD, distance visual acuity, near visual acuity, SKILL score, choroidal thickness, retinal thickness, and retinal volume.

Results: Reticular pseudodrusen was found in 43 participants (41%) on 1 or more imaging method. The SKILL score was significantly worse in those with reticular drusen (mean score ± standard deviation [SD, 38±12) versus those without (mean score ± SD, 33±9) (P = 0.034). Parafoveal retinal thickness, parafoveal retinal volume, and all of the choroidal thickness parameters measured were significantly lower in those with reticular drusen than in those without. The presence of RPD was associated with development of neovascular AMD when corrected for age and sex (odds ratio, 5.5; 95% confidence interval, 1.1–28.8; P = 0.042). All participants in whom geographic atrophy developed during follow-up had visible RPD at baseline.

Conclusions: Significant differences in retinal and choroidal anatomic features, visual function, and risk factor profile exist in unilateral neovascular AMD patients with RPD compared with those without; therefore, such patients should be monitored carefully because of the risk of developing bilateral disease.

Tracking magnetic bright point motions through the solar atmosphere

High-cadence, multiwavelength observations and simulations are employed for the analysis of solar photospheric magnetic bright points (MBPs) in the quiet Sun. The observations were obtained with the Rapid Oscillations in the Solar Atmosphere (ROSA) imager and the Interferometric Bidimensional Spectrometer at the Dunn Solar Telescope. Our analysis reveals that photospheric MBPs have an average transverse velocity of approximately 1 km s-1, whereas their chromospheric counterparts have a slightly higher average velocity of 1.4 km s-1. Additionally, chromospheric MBPs were found to be around 63 per cent larger than the equivalent photospheric MBPs. These velocity values were compared with the output of numerical simulations generated using the muram code. The simulated results were similar, but slightly elevated, when compared to the observed data. An average velocity of 1.3 km s-1 was found in the simulated G-band images and an average of 1.8 km s-1 seen in the velocity domain at a height of 500 km above the continuum formation layer. Delays in the change of velocities were also analysed. Average delays of ˜4 s between layers of the simulated data set were established and values of ˜29 s observed between G-band and Ca ii K ROSA observations. The delays in the simulations are likely to be the result of oblique granular shock waves, whereas those found in the observations are possibly the result of a semi-rigid flux tube.

Dense Multiperson Tracking with Robust Hierarchical Linear Assignment

We introduce a novel dual-stage algorithm for online multi-target tracking in realistic conditions. In the first stage, the problem of data association between tracklets and detections, given partial occlusion, is addressed using a novel occlusion robust appearance similarity method. This is used to robustly link tracklets with detections without requiring explicit knowledge of the occluded regions. In the second stage, tracklets are linked using a novel method of constraining the linking process that removes the need for ad-hoc tracklet linking rules. In this method, links between tracklets are permitted based on their agreement with optical flow evidence. Tests of this new tracking system have been performed using several public datasets.

Adaptive NormalHedge for robust visual tracking

In this paper, we propose a novel visual tracking framework, based on a decision-theoretic online learning algorithm namely NormalHedge. To make NormalHedge more robust against noise, we propose an adaptive NormalHedge algorithm, which exploits the historic information of each expert to perform more accurate prediction than the standard NormalHedge. Technically, we use a set of weighted experts to predict the state of the target to be tracked over time. The weight of each expert is online learned by pushing the cumulative regret of the learner towards that of the expert. Our simulation experiments demonstrate the effectiveness of the proposed adaptive NormalHedge, compared to the standard NormalHedge method. Furthermore, the experimental results of several challenging video sequences show that the proposed tracking method outperforms several state-of-the-art methods.

Efficient tracking of human poses using a manifold hierarchy

In this paper a 3D human pose tracking framework is presented. A new dimensionality reduction method (Hierarchical Temporal Laplacian Eigenmaps) is introduced to represent activities in hierarchies of low dimensional spaces. Such a hierarchy provides increasing independence between limbs, allowing higher flexibility and adaptability that result in improved accuracy. Moreover, a novel deterministic optimisation method (Hierarchical Manifold Search) is applied to estimate efficiently the position of the corresponding body parts. Finally, evaluation on public datasets such as HumanEva demonstrates that our approach achieves a 62.5mm-65mm average joint error for the walking activity and outperforms state-of-the-art methods in terms of accuracy and computational cost.