Object detection is the technique used to identify the location and name of objects. This technology is crucial for the advancement of autonomous driving and related fields. In traditional research, the depth of a stationary object was estimated by assuming the camera movement was known and using changes in the object’s scale. However, this method requires precise knowledge of the camera’s motion in advance. This research estimates the position and velocity of an object in 3D space using a time series of bounding box information calculated for each frame through object detection techniques from images taken with a single camera with unknown camera motion.