28 February 2020 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Photoacoustic tomography (PAT) is a non-invasive imaging modality combining the benefits of optical contrast at ultrasonic resolution. Analytical reconstruction algorithms for photoacoustic signals require large number of data points for accurate image reconstruction. However, in practical scenarios, data is collected using limited number of transducers along with data being often corrupted with noise resulting in only qualitative images. Further, the collected boundary data is band-limited due to limited bandwidth of the transducer making the photoacoustic imaging with limited data being qualitative. In this work, a deep neural network based model with loss function being scaled root-mean-squared-error was proposed for super-resolution, denoising as well as bandwidth enhancement of the photoacoustic signals collected at the boundary of the domain. The proposed network has been compared with traditional as well as other popular deep learning methods in numerical as well as experimental cases and is shown to improve the collected boundary data in turn providing superior quality reconstructed photoacoustic image. The improvement obtained in the Pearson Correlation, Structural Similarity Index Metric and Root Mean Square Error was as high as 35.62%, 33.81% and 41.07% respectively for phantom cases and Signal to Noise Ratio improvement in the reconstructed photoacoustic images was as high as 11.65 dB for in-vivo cases as compared to reconstructed image obtained using original limited bandwidth data. Code is available at https://sites.google.com/site/sercmig/home/dnnpat.
14 April 2023 in Proceedings of the ACM on Human-Computer Interaction
Navigation assistance systems (NASs) aim to help visually impaired people (VIPs) navigate unfamiliar environments. Most of today's NASs support VIPs via turn-by-turn navigation, but a growing body of work highlights the importance of exploration as well. It is unclear, however, how NASs should be designed to help VIPs explore unfamiliar environments. In this paper, we perform a qualitative study to understand VIPs' information needs and challenges with respect to exploring unfamiliar environments to inform the design of NASs that support exploration. Our findings reveal the types of spatial information that VIPs need as well as factors that affect VIPs' information preferences. We also discover specific challenges that VIPs face that future NASs can address, such as orientation and mobility education and collaborating effectively with others. We present design implications for NASs that support exploration, and we identify specific research opportunities and discuss open socio-technical challenges for making such NASs possible. We conclude by reflecting on our study procedure to inform future approaches in research on ethical considerations that may be adopted while interacting with the broader VIP community.
9 September 2020 in IEEE Transactions on Cognitive and Developmental Systems
Abstract– For the past few decades, machines have replaced humans in several disciplines. However, machine cognition still lags behind the human capabilities. We address the machines’ ability to recognize human drawn sketches in this work. Visual representations such as sketches have long been a medium of communication for humans. For artificially intelligent systems to effectively immerse in interactive environments, it is required that machines understand such notations. The abstract nature and varied artistic styling of these sketches make automatic recognition of drawings more challenging than other areas of image classification. In this paper, we use sketches represented as a sequence of strokes, i.e., as vector images, to effectively capture the long-term temporal dependencies in hand-drawn sketches. The proposed approach combines the self-attention capabilities of Transformers while effectively utilizing the long-term temporal dependencies through Temporal Convolution Networks (TCN) for sketch recognition. The confidence scores obtained from the two techniques are combined using triangular-norm (T-norm). Attention heat-maps are plotted to isolate the discriminating parts of a sketch that contribute to sketch classification. The extensive quantitative and qualitative evaluation confirms that the proposed network performs favorably against state-of-the-art techniques.