Vision methods allow for accelerating the measurements of critical parameters of the figure and clothing and making the results free of errors usually made during manual measurements. However, it is necessary to develop appropriate algorithms for estimating spatial parameters, such as circumferences in selected cross-sections, based on 2-dimensional images. The literature is dominated by reports on the use of an ellipse for estimating circumferences, methods involving the concatenation of two ellipses are also described, and examples of the use of a superellipse are discussed but without an in-depth analysis of the selection of its parameters and the results achieved. This article discusses these essential properties of the superellipse as a model of the figure/clothing circumference. The article presents the developed methods of image processing and analysis for measuring circumferences and selected clothing measurement parameters. The paper discusses classic circuit approximation methods using ellipse equations and the proposed superellipse (Lame) method. A vision-based algorithm was presented, which uses spatial markers to measure circumferences and selected clothing parameters. The proposed method was verified on a test stand during experiments and compared with manual measurements. The results of measurements of the circumference, length, and width of selected clothing items were obtained. Statistical estimates characterizing the developed vision measurement system were determined based on the results. The obtained findings indicate that, compared to the classical ellipse approach, in the proposed method to measure the circumference, based on Lame curves, the average absolute error does not exceed 0.55 mm. The average measurement accuracy of the developed vision system for all conducted measurements was below the assumed value of 5 mm.