Motion Segmentation & Multiple Object Tracking by Correlation Co-Clustering.

Models for computer vision are commonly defined either w.r.t. low-level concepts such as pixels that are to be grouped, or w.r.t. high-level concepts such as semantic objects that are to be detected and tracked. Combining bottom-up grouping with top-down detection and tracking, although highly desirable, is a challenging problem. We state this joint problem as a co-clustering problem that is principled and tractable by existing algorithms. We demonstrate the effectiveness of this approach by combining bottom-up motion segmentation by grouping of point trajectories with high-level multiple object tracking by clustering of bounding boxes. We show that solving the joint problem is beneficial at the low-level, in terms of the FBMS59 motion segmentation benchmark, and at the high-level, in terms of the Multiple Object Tracking benchmarks MOT15, MOT16 and the MOT17 challenge, and is state-of-the-art in some metrics.

Iron deficiency is prevalent in a sample of pregnant women at delivery in Germany.

OBJECTIVE: To determine the prevalence and risk factors of iron deficiency in pregnancy, since iron supplements are not routine in Germany. STUDY DESIGN: Soluble transferrin receptor (sTfR), ferritin, hemoglobin (Hb), C-reactive protein (CRP) and leucocyte counts were determined in venous blood samples of 378 women before delivery; 191 of them filled in a questionnaire. Statistical analysis was performed using SPSS 9.0.1. RESULTS: CRP and leucocyte count correlated significantly with ferritin values, while sTfR values were independent. Iron deficiency (sTfR >3.3 mg/dl) was found in 40.7% of pregnant women, and anemia (Hb<11.0 g/dl) in 13.6%. Non-German nationality, low educational level and young maternal age were significant risk factors for iron deficiency, while alcohol use in pregnancy, iron and folic acid supplements were protective. After adjusting for all other factors in a logistic regression equation, not taking any iron supplements in pregnancy (odd ratios (OR) 3.3 and 95% confidence interval (CI) 1.1-9.4), and young maternal age (OR 2.86 and 95% CI 1.1-7.7) remained significant risks. CONCLUSION: Iron deficiency in pregnant women in Germany is prevalent; it could be prevented by routine iron supplements.