A method for estimating 6-DOF relative displacement using vision-based localization and apparatus therefor are disclosed. A method for estimating 6-DOF relative displacement may include acquiring images of a first marker attached to a fixing member and a second marker attached to a dynamic member for assembling to the fixing member by using a camera, extracting a feature point of the first marker and a feature point of the second marker through image processing for the acquired images, and estimating 6-DOF relative displacement of the dynamic member for the fixing member based on the extracted feature point of the first marker and feature point of the second marker.