Recent crowd counting approaches have achieved excellent performance. However, they are essentially based on fully supervised paradigm and require large number of annotated samples. Obtaining annotations is an expensive and labour-intensive process. In this work, we focus on reducing the annotation efforts by learning to count in the crowd from limited number of labeled samples while leveraging a large pool of unlabeled data. Specifically, we propose a Gaussian Process-based iterative learning mechanism that involves estimation of pseudo-ground truth for the unlabeled data, which is then used as supervision for training the network. The proposed method is shown to be effective under the reduced data (semi-supervised) settings for several datasets like ShanghaiTech, UCF-QNRF, WorldExpo, UCSD, etc. Furthermore, we demonstrate that the proposed method can be leveraged to enable the network in learning to count from synthetic dataset while being able to generalize better to real-world datasets (synthetic-to-real transfer).
Titanium oxide (TiO 2 ) films were deposited on silicon (100) and quartz substrates at various substrate temperatures (300 -873 K) at an optimized oxygen partial pressure of 3.0 × 10 −2 mbar by pulsed laser deposition. The effect of substrate temperature on structure, surface morphology and optical properties of the films were investigated using X-ray diffraction (XRD), atomic force microscopy (AFM) and photoluminescence spectroscopy (PL) respectively. The XRD results showed that the films are polycrystalline in nature and have tetragonal structure. The film prepared at higher substrate temperature showed strong rutile phase. The results indicated that all the films possess both phases (anatase and rutile) of titania. The AFM shows the crystalline nature, dense, uniform distribution of the nanocrystallites with a surface roughness of 2 -8 nm. The photoluminescence studies showed the asymmetric peak ∼ 370 nm indicating the bandgap for the TiO 2 films.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.