Feature tracking is a key, underlying component in many approaches to 3D reconstruction, detection, localization and recognition of underwater objects. In this paper, we proposed to adapt SIFT technique for feature tracking in underwater video sequences. Over the past few years the underwater vision is attracting researchers to investigate suitable feature tracking techniques for underwater applications. The researchers have developed many feature tracking techniques such as KLT, SIFT, SURF etc., to track the features in video sequence for general applications. The literature survey reveals that there is no standard feature tracker suitable for underwater environment. We proposed to adapt SIFT technique for tracking features of objects in underwater video sequence. The SIFT extracts features, which are invariant to scale, rotation and affine transformations. We have compared and evaluated SIFT with popular techniques such as KLT and SURF on captured video sequence of underwater objects. The experimental results shows that adapted SIFT works well for underwater video sequence.