With recent progress in graphics, it has become more tractable to train models on synthetic images, potentially avoiding the need for expensive annotations. However, learning from synthetic images may not achieve the desired performance due to a gap between synthetic and real image distributions. To reduce this gap, we propose Simulated+Unsupervised (S+U) learning, where the task is to learn a model to improve the realism of a simulator's output using unlabeled real data, while preserving the annotation information from the simulator. We develop a method for S+U learning that uses an adversarial network similar to Generative Adversarial Networks (GANs), but with synthetic images as inputs instead of random vectors. We make several key modifications to the standard GAN algorithm to preserve annotations, avoid artifacts, and stabilize training: (i) a 'self-regularization' term, (ii) a local adversarial loss, and (iii) updating the discriminator using a history of refined images. We show that this enables generation of highly realistic images, which we demonstrate both qualitatively and with a user study. We quantitatively evaluate the generated images by training models for gaze estimation and hand pose estimation. We show a significant improvement over using synthetic images, and achieve state-of-the-art results on the MPIIGaze dataset without any labeled real data.

No abstract

In this paper, we propose a multiple kernel learning (MKL) algorithm that is based on the sparse representation-based classification (SRC) method. Taking advantage of the nonlinear kernel SRC in efficiently representing the nonlinearities in the high-dimensional feature space, we propose an MKL method based on the kernel alignment criteria. Our method uses a two step training method to learn the kernel weights and sparse codes. At each iteration, the sparse codes are updated first while fixing the kernel mixing coefficients, and then the kernel mixing coefficients are updated while fixing the sparse codes. These two steps are repeated until a stopping criteria is met. The effectiveness of the proposed method is demonstrated using several publicly available image classification databases and it is shown that this method can perform significantly better than many competitive image classification algorithms.

While recent techniques for discriminative dictionary learning have demonstrated tremendous success in image analysis applications, their performance is often limited by the amount of labeled data available for training. Even though labeling images is difficult, it is relatively easy to collect unlabeled images either by querying the web or from public datasets. In this paper, we propose a discriminative dictionary learning technique which utilizes both labeled and unlabeled data for learning dictionaries. Extensive evaluation on existing datasets demonstrate that the proposed method performs significantly better than state of the art dictionary learning approaches when unlabeled images are available for training.

This study deals with the analysis and implementation of an HPF (high power factor) single-stage, single switch buck converter-based power supply design for an LED (light emitting diode) lamp load of 13 W operated at the universal ac mains. In general purpose lighting applications, a buck converter is a good candidate for power factor correction with low component count and reduced cost. In low-power lighting, it is a tough task to control the THD i (total harmonic distortion) of ac mains current under the limits of strict international standards such as IEC-61000-3-2 with universal ac mains for class D equipments. In the proposed optocouplerless topology, HPF operation at input ac mains is achieved by operating the buck ac-dc converter in continuous conduction mode. The design, modelling and simulation of the proposed topology are executed using MATLAB-Simulink and sim-power system toolboxes. A prototype of the power supply for LED lamp is developed for multiple LEDs connected in series configuration. The efficiency of the proposed LED lamp driver is observed as 83.76% at rated voltage of 220 V and the THD of ac mains current less than 17.27% for a wide range of voltages of 90-270 V.