No abstract

Visual appearance-based person retrieval is a challenging problem in surveillance. It uses attributes like height, cloth color, cloth type and gender to describe a human. Such attributes are known as soft biometrics. This paper proposes person retrieval from surveillance video using height, torso cloth type, torso cloth color and gender. The approach introduces an adaptive torso patch extraction and bounding box regression to improve the retrieval. The algorithm uses fine-tuned Mask R-CNN and DenseNet-169 for person detection and attribute classification respectively. The performance is analyzed on AVSS 2018 challenge II dataset and it achieves 11.35% improvement over state-of-the-art based on average Intersection over Union measure.

QR codes have become widely popular along with the increased usage and popularity of smart phones. In many areas, QR codes have overtaken the place of linear barcodes because of the obvious advantage of large data capacity and ease of data retrieval. QR code specifications offers many more advantages like reduced space, durability against soil and damage, high data capacity, supported languages are more than other barcodes, supports 360 degree reading, etc over linear barcodes which makes QR codes worth opting. The structural flexibility of QR code architecture opens many more possibilities to stretch the limits of data capacity further away which includes data hiding techniques, multiplexing techniques, use of color QR codes, use of data compression techniques, etc. Proposed technique attempts to increase data capacity by multiplexing several QR codes and generating a color QR code. Experimental results show that this technique offers increase of the data capacity upto 24 times as compared to a standard QR code of same size. To get quicker results while demultiplexing, multiplexing of 12 or less QR codes is advisable with proposed technique. Due to such large capacity offered by proposed technique, embedding of speech signal into a QR code has made possible.

No abstract

No abstract