SeasonDepth: Cross-Season Monocular Depth Prediction Dataset and Benchmark under Multiple Environments

Changing environments poses a great challenge on the outdoor visual perception and scene understanding for robust long-term autonomous driving and mobile robots, where depth-auxiliary geometric information plays an essential role to the robustness under challenging scenes. Although monocular depth prediction has been well studied recently, there are few work focusing on the depth prediction across multiple environmental conditions, e.g. changing illumination and seasons, owing to the lack of such a real-world dataset and benchmark. In this work, a new cross-season monocular depth prediction dataset SeasonDepth (available on https://seasondepth.github.io) is derived from CMU Visual Localization dataset through structure from motion. To benchmark the depth estimation performance under different environments, we investigate representative and recent state-of-the-art open-source supervised, self-supervised and domain adaptation depth prediction methods from KITTI benchmark using several newly-formulated metrics. Through extensive experimental evaluation on the proposed dataset without fine-tuning, the influence of multiple environments on performance and robustness is analyzed both qualitatively and quantitatively, showing that the long-term monocular depth prediction is far from solved. We further give promising solutions especially with stereo geometry and multi-task sequential self-supervised training to enhance the robustness to changing environments.