This is the supplementary video of our paper "Multi-Scale Surface Normal Estimation from Depth Maps", which is accepted for publication in IMPROVE 2023 and has been nominated for the Best Student Paper Award.
In this study we have developed a simple yet effective algorithm that estimates the surface normal vectors from complete and incomplete depth maps. As it can be seen in this video our method can estimate the surface normals for real-world-like scenes effectively.
If you use our work in your research please cite the following:
Ulucan, D., Ulucan, O. and Ebner, M. (2023). Multi-Scale Surface Normal Estimation from Depth Maps. In International Conference on Image Processing and Vision Engineering, pages 45-54, Prague, Czechia, Science and Technology Publications.
This video is created by using the frames in the New Tsukuba Dataset (Martull et al., 2012, Peris et al., 2012).
[1] Martull, S., Peris, M., and Fukui, K. (2012). Realistic cg stereo image dataset with ground truth disparity maps. In ICPR Workshop TrakMark2012, volume 111, pages 117–118, Tsukuba, Japan, IEEE.
[2] Peris, M., Martull, S., Maki, A., Ohkawa, Y., and Fukui, K. (2012). Towards a simulation driven stereo vision system. In ICPR, pages 1038-1042, Tsukuba, Japan, IEEE.
