Fig. 6

Comparison of the different methods on the Temple, City1, and City2 datasets. In all [(a), (b) and (c)], the top row, from left to right, are the reconstruction results of our proposed method, Global, GDMR, and FSSR, respectively. The next row contains two details of the surface, and the columns in the two details represent the result of our method, Global, GDMR, and FSSR, respectively. The red circles highlight the surface quality of our approach, demonstrating that our proposed method and Global can obtain similar results. Both methods outperform GDMR and FSSR in surface details and completeness