Search for: All records

Self-supervised depth estimation has recently demonstrated promising performance compared to the supervised methods on challenging indoor scenes. However, the majority of efforts mainly focus on exploiting photometric and geometric consistency via forward image warping and backward image warping, based on monocular videos or stereo image pairs. The influence of defocus blur to depth estimation is neglected, resulting in a limited performance for objects and scenes in out of focus. In this work, we propose the first framework for simultaneous depth estimation from a single image and image focal stacks using depth-from-defocus and depth-from-focus algorithms. The proposed network is able to learn optimal depth mapping from the information contained in the blur of a single image, generate a simulated image focal stack and all-in-focus image, and train a depth estimator from an image focal stack. In addition to the validation of our method on both synthetic NYUv2 dataset and real DSLR dataset, we also collect our own dataset using a DSLR camera and further verify on it. Experiments demonstrate that our system surpasses the state-of-the-art supervised depth estimation method over 4% in accuracy and achieves superb performance among the methods without direct supervision on the synthesized NYUv2 dataset, which has been rarely explored. 

From the discovery of functionally graded laminated composites, to near-structurally optimized diagonally reinforced square lattice structures, the skeletal system of the predominantly deep-sea sponge Euplectella aspergillum has continued to inspire biologists, materials scientists and mechanical engineers. Building on these previous efforts, in the present study, we develop an integrated finite element and fluid dynamics approach for investigating structure–function relationships in the complex maze-like organization of helical ridges that surround the main skeletal tube of this species. From these investigations, we discover that not only do these ridges provide additional mechanical reinforcement, but perhaps more significantly, provide a critical hydrodynamic benefit by effectively suppressing von Kármán vortex shedding and reducing lift forcing fluctuations over a wide range of biologically relevant flow regimes. By comparing the disordered sponge ridge geometry to other more symmetrical strake-based vortex suppression systems commonly employed in infrastructure applications ranging from antennas to underwater gas and oil pipelines, we find that the unique maze-like ridge organization of E. aspergillum can completely suppress vortex shedding rather than delaying their shedding to a more downstream location, thus highlighting their potential benefit in these engineering contexts.