Deep Photometric Stereo Network with Multi-Scale Feature Aggregation

We present photometric stereo algorithms robust to non-Lambertian reflection, which are based on a convolutional neural network in which surface normals of objects with complex geometry and surface reflectance are estimated from a given set of an arbitrary number of images. These images are taken from the same viewpoint under different directional illumination conditions. The proposed method focuses on surface normal estimation, where multi-scale feature aggregation is proposed to obtain a more accurate surface normal, and max pooling is adopted to obtain an intermediate order-agnostic representation in the photometric stereo scenario. The proposed multi-scale feature aggregation scheme using feature concatenation is easily incorporated into existing photometric stereo network architectures. Our experiments were performed with a DiLiGent photometric stereo benchmark dataset consisting of ten real objects, and they demonstrated that the accuracies of our calibrated and uncalibrated photometric stereo approaches were improved over those of baseline methods. In particular, our experiments also demonstrated that our uncalibrated photometric stereo outperformed the state-of-the-art method. Our work is the first to consider the multi-scale feature aggregation in photometric stereo, and we showed that our proposed multi-scale fusion scheme estimated the surface normal accurately and was beneficial to improving performance.

Comparison of stochastic parameterizations in the framework of a coupled ocean–atmosphere model

A new framework is proposed for the evaluation of stochastic subgrid-scale parameterizations in the context of the Modular Arbitrary-Order Ocean-Atmosphere Model (MAOOAM), a coupled ocean–atmosphere model of intermediate complexity. Two physically based parameterizations are investigated – the first one based on the singular perturbation of Markov operators, also known as homogenization. The second one is a recently proposed parameterization based on Ruelle's response theory. The two parameterizations are implemented in a rigorous way, assuming however that the unresolved-scale relevant statistics are Gaussian. They are extensively tested for a low-order version known to exhibit low-frequency variability (LFV), and some preliminary results are obtained for an intermediate-order version. Several different configurations of the resolved–unresolved-scale separations are then considered. Both parameterizations show remarkable performances in correcting the impact of model errors, being even able to change the modality of the probability distributions. Their respective limitations are also discussed.

Synthesis and structures of six closely related N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]arylamides, together with an isolated reaction intermediate: order versus disorder, molecular conformations and hydrogen bonding in zero, one and two dimensions

Six closely related N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]arylamides have been synthesized and structurally characterized, together with a representative reaction intermediate. In each of N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]benzamide, C20H16ClNO2S, (I), N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]-4-phenylbenzamide, C26H20ClNO2S, (II), and 2-bromo-N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]benzamide, C20H15BrClNO2S, (III), the molecules are disordered over two sets of atomic sites, with occupancies of 0.894 (8) and 0.106 (8) in (I), 0.832 (5) and 0.168 (5) in (II), and 0.7006 (12) and 0.2994 (12) in (III). In each of N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]-2-iodobenzamide, C20H15ClINO2S, (IV), and N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]-2-methoxybenzamide, C21H18ClNO3S, (V), the molecules are fully ordered, but in N-[3-(2-chlorobenzoyl)-5-ethylthiophen-2-yl]-2,6-difluorobenzamide, C20H14ClF2NO2S, (VI), which crystallizes with Z′ = 2 in the space group C2/c, one of the two independent molecules is fully ordered, while the other is disordered over two sets of atomic sites having occupancies of 0.916 (3) and 0.084 (3). All of the molecules in compounds (I)–(VI) exhibit an intramolecular N—H...O hydrogen bond. The molecules of (I) and (VI) are linked by C—H...O hydrogen bonds to form finite zero-dimensional dimers, which are cyclic in (I) and acyclic in (VI), those of (III) are linked by C—H...π(arene) hydrogen bonds to form simple chains, and those of (IV) and (V) are linked into different types of chains of rings, built in each case from a combination of C—H...O and C—H...π(arene) hydrogen bonds. Two C—H...O hydrogen bonds link the molecules of (II) into sheets containing three types of ring. In benzotriazol-1-yl 3,4-dimethoxybenzoate, C15H13N3O4, (VII), the benzoate component is planar and makes a dihedral angle of 84.51 (6)° with the benzotriazole unit. Comparisons are made with related compounds.

Comparison of stochastic parameterizations in the framework of a coupled ocean-atmosphere model

A new framework is proposed for the evaluation of stochastic subgrid-scale parameterizations in the context of MAOOAM, a coupled ocean-atmosphere model of intermediate complexity. Two physically-based parameterizations are investigated, the first one based on the singular perturbation of Markov operator, also known as homogenization. The second one is a recently proposed parameterization based on the Ruelle's response theory. The two parameterization are implemented in a rigorous way, assuming however that the unresolved scale relevant statistics are Gaussian. They are extensively tested for a low-order version known to exhibit low-frequency variability, and some preliminary results are obtained for an intermediate-order version. Several different configurations of the resolved-unresolved scale separations are then considered. Both parameterizations show remarkable performances in correcting the impact of model errors, being even able to change the modality of the probability distributions. Their respective limitations are also discussed.

Almost sure limit theorem for the order statistics of stationary Gaussian sequences

In this paper, by using a new comparison inequality for order statistics of Gaussian variables, we proved an almost sure central limit theorem for extreme order statistics of stationary Gaussian sequences with covariance rn under the condition rn log n(log log n)1+? = O(1) for some ? > 0. A similar result on intermediate order statistics is also proved for stationary Gaussian sequences. The obtained results improve some of the existing results.