Evaluation of CNN-based Single-Image Depth Estimation Methods
This work addresses the problem of limited evaluation metrics for researchers and practitioners in computer vision, though it is incremental as it builds on existing depth estimation methods.
The authors tackled the lack of detailed evaluation schemes for CNN-based single-image depth estimation methods by proposing novel quality criteria focusing on edges, planar regions, depth consistency, and absolute accuracy, and they introduced a new high-quality RGB-D dataset using a DSLR camera and laser scanner to validate their protocol.
While an increasing interest in deep models for single-image depth estimation methods can be observed, established schemes for their evaluation are still limited. We propose a set of novel quality criteria, allowing for a more detailed analysis by focusing on specific characteristics of depth maps. In particular, we address the preservation of edges and planar regions, depth consistency, and absolute distance accuracy. In order to employ these metrics to evaluate and compare state-of-the-art single-image depth estimation approaches, we provide a new high-quality RGB-D dataset. We used a DSLR camera together with a laser scanner to acquire high-resolution images and highly accurate depth maps. Experimental results show the validity of our proposed evaluation protocol.