• Sharon Priyanka P
• Satish B Basapur

Checking the number of individuals present in the gathering is an extraordinary mission, this task takes a more extensive view to address group tallying from the semantic demonstrating angle. Basically, group checking is a test of passerby semantic assessment including 3 key elements: heads, pedestrians, and their context structure. The records of different casing segments is a basic sign to enable us to anticipate whether there exists an individual at a specific point. Existing methodologies, perform counting of group from the demeanor of straightforwardly displaying the visual homes of either the heads or the total body or, beyond clearly catching the blended edge part semantic shape actualities which is essential for group checking. In this accession, we consider semantic scene models to define the significant components of group. Then, change the group checking hassle directly into a multi-task becoming more acquainted with inconvenience, to such an extent that the semantic scene variants are changed into distinct multi-errands. At last, the deep convolutional neural systems are utilized to examine the sub-assignments in a unified scheme. This technique encodes the semantic idea of group counting and gives a novel answer regarding semantic analysis of person on foot. The semantic shape records is exhibited to be a ground-breaking sign in scene of group counting.

Deep Convolutional Neural Network, Group Counting, Multi-errands, Semantic Scene Model, Visual Context Structure

"Group counting using Nets", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.330-332, June 2019, Available :http://www.jetir.org/papers/JETIR1906P51.pdf

"Group counting using Nets", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp330-332, June 2019, Available at : http://www.jetir.org/papers/JETIR1906P51.pdf