Challenges of Depth Estimation for Transparent Objects

Transparent objects and surfaces are pervasive in man-made environments and need to be considered in any vision system. Accurate depth data is a key factor for the reliability of such systems, requiring methods tailored for transparency to overcome the sensing shortcomings. However, the current state-of-the-art methods to predict the depth of such objects are not yet reliable enough to ensure safe operation of autonomous systems in arbitrary complex environments. In order to better understand and improve upon existing solutions, we evaluate the performance of a variety of depth estimation methods. Doing so, we disentangle the different factors impacting their performance. Among our findings, neural radiance fields offer the best accuracy, but are very sensitive to the number of images used to understand the scene, and do not benefit from any level of object understanding to help them fill in the gaps.