Single-Image Depth Perception in the Wild
This work addresses depth estimation from single images in real-world scenarios, which is incremental as it builds on existing methods with new data and algorithmic tweaks.
The paper tackles single-image depth perception in unconstrained settings by introducing a new dataset with relative depth annotations and a simpler algorithm that outperforms state-of-the-art methods, showing significant improvements in performance.
This paper studies single-image depth perception in the wild, i.e., recovering depth from a single image taken in unconstrained settings. We introduce a new dataset "Depth in the Wild" consisting of images in the wild annotated with relative depth between pairs of random points. We also propose a new algorithm that learns to estimate metric depth using annotations of relative depth. Compared to the state of the art, our algorithm is simpler and performs better. Experiments show that our algorithm, combined with existing RGB-D data and our new relative depth annotations, significantly improves single-image depth perception in the wild.