AcED: Accurate and Edge-consistent Monocular Depth Estimation
This work addresses the challenge of accurate depth estimation from monocular images, which is incremental by improving upon existing ordinal regression approaches.
The paper tackled the problem of single image depth estimation by proposing a fully differentiable ordinal regression method trained end-to-end, resulting in smooth and edge-consistent depth maps with demonstrated superiority over state-of-the-art methods on benchmarks.
Single image depth estimation is a challenging problem. The current state-of-the-art method formulates the problem as that of ordinal regression. However, the formulation is not fully differentiable and depth maps are not generated in an end-to-end fashion. The method uses a naïve threshold strategy to determine per-pixel depth labels, which results in significant discretization errors. For the first time, we formulate a fully differentiable ordinal regression and train the network in end-to-end fashion. This enables us to include boundary and smoothness constraints in the optimization function, leading to smooth and edge-consistent depth maps. A novel per-pixel confidence map computation for depth refinement is also proposed. Extensive evaluation of the proposed model on challenging benchmarks reveals its superiority over recent state-of-the-art methods, both quantitatively and qualitatively. Additionally, we demonstrate practical utility of the proposed method for single camera bokeh solution using in-house dataset of challenging real-life images.