Hi!
I'm Dinesh

Abstract:

In this paper, we propose an image matting framework called Salient Image Matting to estimate the per-pixel opacity value of the most salient foreground in an image. To deal with a large amount of semantic diversity in images, a trimap is conventionally required as it provides important guidance about object semantics to the matting process. However, creating a good trimap is often expensive and timeconsuming. The SIM framework simultaneously deals with the challenge of learning a wide range of semantics and salient object types in a fully automatic and an end to end manner. Specifically, our framework is able to produce accurate alpha mattes for a wide range of foreground objects and cases where the foreground class, such as human, appears in a very different context than the train data directly from an RGB input. This is done by employing a salient object detection model to produce a trimap of the most salient object in the image in order to guide the matting model about higher-level object semantics. Our framework leverages large amounts of coarse annotations coupled with a heuristic trimap generation scheme to train the trimap prediction network so it can produce trimaps for arbitrary foregrounds. Moreover, we introduce a multi-scale fusion architecture for the task of matting to better capture finer, low-level opacity semantics. With high-level guidance provided by the trimap network, our framework requires only a fraction of expensive matting data as compared to other automatic methods while being able to produce alpha mattes for a diverse range of inputs. We demonstrate our framework on a range of diverse images and experimental results show our framework compares favourably against state of art matting methods without the need for a trimap.

Link to Paper

Abstract:

5G is the next wave in the communication industry where the end customers enjoy multiple services with minimum latency. Since 5G towers operate in high-frequency spectrum the coverage of these towers is very limited when compared with its earlier versions and this facilitates the requirement of installing additional number of mobile towers. One of the key requirements for enabling the full-scale deployment of the new towers in addition to the existing mobile towers are to ensure their regular inspection and maintenance without human intervention. While there has been research carried out to develop autonomous drones for monitoring mobile towers, the development is limited to the inspection stage and not capable of fixing the faults without human assistance. The fundamental limitation that prevented carrying out a repairing task was the absence of a competent robotic arm that can act as a human arm. Traditional drones are only equipped with a robotic manipulator on the bottom using which cannot access all the points in a typical wireless infrastructure with a positive slope. This paper designs an autonomous drone, from scratch, with a robotic arm on the top to inspect and repair faults in mobile towers. The main contribution is a novel coupling algorithm that takes proactive action to stabilize the drone while the arm is carrying out the mending tasks with less energy. The drone is also tested in a real- time windy environment by emulating the mobile tower scenario.

Link to Paper

Abstract:

Most automobiles lack reliable smart systems that can constantly track the driver's behaviour and raise alarms as required. Extant systems are either too slow or not robust enough to cope with different types of drivers and conditions. In this paper, a robust system to continuously track the driver's eye and detect its state (open/close) is proposed. Frames from a live camera feed are constantly processed. Viola Jones algorithm, using Haar filters extracts the eye. The extraction is efficient with and without spectacles (translucent) and the system can even estimate the Region of Interest (RoI) where it is most likely to find the eye in the event that no eyes are explicitly detected. A trained CNN model using the LeNet architecture classifies the extracted eyes. The rate at which predictions are made is also higher than existing systems. The system raises an alarm if, after analysing the data points, it detects any anomalies.

Link to Paper

Abstract:

Deep learning and other big data technologies have over time become very powerful and accurate. There are algorithms and models developed that have near human accuracy in their task. In health care, the amount of data available is massive and hence, these technologies have a great scope. This paper reviews a few interesting contributions to the field specifically to medical imaging, genomics and patient health records.

Link to Paper