Learning to Reconstruct and Understand Indoor Scenes From Sparse Views.

This paper proposes a new method for simultaneous 3D reconstruction and semantic segmentation for indoor scenes. Unlike existing methods that require recording a video using a color camera and/or a depth camera, our method only needs a small number of (e.g., 3~5) color images from uncalibrated sparse views, which significantly simplifies data acquisition and broadens applicable scenarios. To achieve promising 3D reconstruction from sparse views with limited overlap, our method first recovers the depth map and semantic information for each view, and then fuses the depth maps into a 3D scene. To this end, we design an iterative deep architecture, named IterNet, to estimate the depth map and semantic segmentation alternately. To obtain accurate alignment between views with limited overlap, we further propose a joint global and local registration method to reconstruct a 3D scene with semantic information. We also make available a new indoor synthetic dataset, containing photorealistic high-resolution RGB images, accurate depth maps and pixel-level semantic labels for thousands of complex layouts. Experimental results on public datasets and our dataset demonstrate that our method achieves more accurate depth estimation, smaller semantic segmentation errors, and better 3D reconstruction results over state-of-the-art methods.

Visualising reactive oxygen species in live mammals and revealing of ROS-related system.

Of all the aerobic respiration by-products, cytotoxic superoxide derived from mitochondrial-leaked electrons, is the only one known to be disposed of intracellularly. Is this fate the only destiny for mitochondrial-leaked electrons? When Cynomolgus monkeys were injected intravenously with reactive oxygen species (ROS) indicators, the connective tissues of dura mater, facial fascia, pericardium, linea alba, dorsa fascia and other body parts, emitted specific and intense fluorescent signals. Moreover, the fluorescent signals along the linea alba of SD rats, did not result from the local presence of ROS but from the interaction of ROS indicators with electrons flowing through this tissue. Furthermore, the electrons travelling along the linea alba of mice were revealed to originate from mitochondria. These data suggest that mitochondrial-leaked electrons may be transported extracellularly to a hitherto undescribed system of connective tissues, which is pervasively networked, electrically conductive and metabolically related.