3D in natural random refractive distortions

Abstract

Random distortions naturally affect images taken through atmospheric turbulence or wavy water. They pose new 3D recovery problems. Distortions are caused by the volumetric field of turbulent air or the 3D shape of water waves. We show methods that recover these 3D distorting media. Moreover, it is possible to triangulate objects beyond the refracting medium. Applications include sensing and study of random refractive media in nature, and enhanced imaging including possibilities for a virtual periscope.

Multi-view stereo of an object immersed in a refractive medium

“…The case of a camera inside an underwater dome is studied in [35]. More complex situations have explored when seeing aerial objects from underwater through a randomly moving surface in [36,37]. The converse situation (air to water) is explored in [38] using a differentiable/NN framework.…”

“…Scenarios like viewing aerial objects from underwater are covered in Refs. 30 and 31 and air to water transitions in Ref. 32.…”

Water-Air Interface Imaging: Recovering the Images Distorted by Surface Waves via an Efficient Registration Algorithm

“…Moreover, an optimization strategy which combines the JADE with LBFGS methods is introduced to accelerate the iterative registration process. Unlike other methods, ours does not need a particular camera height [ 19 ], image priori [ 20 ], or special illumination [ 36 , 37 ]. Our approach is similar to the state-of-the-art [ 25 , 28 , 29 , 30 ] in that it only needs a short sequence (61 frames) rather than 800 frames in [ 14 ], 300k frames in [ 22 ], 100 frames in [ 23 ], and 43,600 frames in [ 24 ], where it is the prerequisite for underwater imaging systems to achieve covert observation.…”