According to one embodiment, a plurality of moving objects is detected from a plurality of frames acquired in time series. Each of the moving objects is corresponded among the frames. A tracklet of each moving object corresponded is extracted and stored. A frame to calculate a position of a moving object is set to a notice frame. The frames are grouped into a first block including at least the notice frame, a second block positioned before the first block in time series, and a third block positioned after the first block in time series. A secondary tracklet included in the second block is acquired from the stored tracklets. The secondary tracklet is corresponded with tracklets included in the first block and the third block, based on a similarity between the secondary tracklet and each of the tracklets. The secondary tracklet is associated with the corresponded tracklets, as a tertiary tracklet.