Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer

🤖 Plain-English Summary

The success of monocular depth estimation relies on large and diverse training sets. The experiments confirm that mixing data from complementary sources greatly improves monocular depth estimation.

🔑 Key Findings

• Due to the challenges associated with acquiring dense ground-truth depth across different environments at scale, a number of datasets with distinct characteristics and biases have emerged.
• We develop tools that enable mixing multiple datasets during training, even if their annotations are incompatible.
• In particular, we propose a robust training objective that is invariant to changes in depth range and scale, advocate the use of principled multi-objective learning to combine data from different sources, and highlight the importance of pretraining encoders on auxiliary tasks.

🔗 Related Resources