Convex hull, a minimum enclosing convex envelope is a popular construction in the field of computational geometry. Typical convex hull include not just the points in the hull but also the connectivity between them. However, at times, it may be sufficient to get only points contributing to the hull, which can then be employed in constructions such as positive α-hull. Algorithms for convex hull primarily focus on point-set as input. Very few algorithms handle input curves as such, without discretization. In this paper, algorithm for determining curves that belong to convex hull using minimum spanning tree (MST) is proposed. The edges of the MST can be considered as a rubber-band connecting all the curves with zero enclosed area. Valency of nodes in the MST is used to define curve triplets. Maximum inscribed circle (MIC) for all triplets is identified (and stored in a triplet matrix) and its minimum is then picked to identify the starting triplet. The triplet is then deleted, updating the triplet matrix. The deletion of the triplet is akin to leaving out curves in the interior to the convex hull. Updating a triplet matrix emulates the pushing out the rubber-band (edges of MST) and moving towards the curves in the convex hull. Repeating the process of deleting a triplet and updating the triplet matrix will eventually form pairs of curves that lie in the convex hull. Connectivity information is then derived using re-ordering. Results are presented that show curves belonging to convex hull. © 2013 Copyright CAD Solutions, LLC.