多分辨率的深度学习方法，例如U-NET体系结构，在分类和分割图像中已经达到了高性能。但是，这些方法不能提供潜在的图像表示形式，也不能用于分解，denoise和重建图像数据。 U-NET和其他卷积神经网络（CNNS）通常使用合并来扩大接受场，这通常会导致不可逆的信息丢失。这项研究建议包括riesz-quincunx（RQ）小波变换，结合1）高阶Riesz小波变换和2）在U-NET体系结构内正交Quincunx小波（两者都用于减少医学图像中的模糊） ，以减少卫星图像及其时间序列中的噪音。在变换的特征空间中，我们提出了一种变异方法，以了解特征的随机扰动如何影响图像以进一步降低噪声。结合两种方法，我们引入了一种用于减少卫星图像中噪声的图像和时间序列分解的混合Rqunet-VAE方案。我们提出了定性和定量的实验结果，表明与其他最先进的方法相比，我们提出的Rqunet-VAE在降低卫星图像中的噪声方面更有效。我们还将我们的方案应用于多波段卫星图像的多个应用程序，包括：通过扩散和图像分割分解图像denoising，图像和时间序列分解。

Reliable and cost-effective counting of people in large indoor spaces is a significant challenge with many applications. An emerging approach is to deploy multiple fisheye cameras mounted overhead to monitor the whole space. However, due to the overlapping fields of view, person re-identificaiton (PRID) is critical for the accuracy of counting. While PRID has been thoroughly researched for traditional rectilinear cameras, few methods have been proposed for fisheye cameras and their performance is comparatively lower. To close this performance gap, we propose a multi-feature framework for fisheye PRID where we combine deep-learning, color-based and location-based features by means of novel feature fusion. We evaluate the performance of our framework for various feature combinations on FRIDA, a public fisheye PRID dataset. The results demonstrate that our multi-feature approach outperforms recent appearance-based deep-learning methods by almost 18% points and location-based methods by almost 3% points in accuracy.

Heteroscedastic regression models a Gaussian variable's mean and variance as a function of covariates. Parametric methods that employ neural networks for these parameter maps can capture complex relationships in the data. Yet, optimizing network parameters via log likelihood gradients can yield suboptimal mean and uncalibrated variance estimates. Current solutions side-step this optimization problem with surrogate objectives or Bayesian treatments. Instead, we make two simple modifications to optimization. Notably, their combination produces a heteroscedastic model with mean estimates that are provably as accurate as those from its homoscedastic counterpart (i.e.~fitting the mean under squared error loss). For a wide variety of network and task complexities, we find that mean estimates from existing heteroscedastic solutions can be significantly less accurate than those from an equivalently expressive mean-only model. Our approach provably retains the accuracy of an equally flexible mean-only model while also offering best-in-class variance calibration. Lastly, we show how to leverage our method to recover the underlying heteroscedastic noise variance.

The identification of material parameters occurring in constitutive models has a wide range of applications in practice. One of these applications is the monitoring and assessment of the actual condition of infrastructure buildings, as the material parameters directly reflect the resistance of the structures to external impacts. Physics-informed neural networks (PINNs) have recently emerged as a suitable method for solving inverse problems. The advantages of this method are a straightforward inclusion of observation data. Unlike grid-based methods, such as the finite element method updating (FEMU) approach, no computational grid and no interpolation of the data is required. In the current work, we aim to further develop PINNs towards the calibration of the linear-elastic constitutive model from full-field displacement and global force data in a realistic regime. We show that normalization and conditioning of the optimization problem play a crucial role in this process. Therefore, among others, we identify the material parameters for initial estimates and balance the individual terms in the loss function. In order to reduce the dependence of the identified material parameters on local errors in the displacement approximation, we base the identification not on the stress boundary conditions but instead on the global balance of internal and external work. In addition, we found that we get a better posed inverse problem if we reformulate it in terms of bulk and shear modulus instead of Young's modulus and Poisson's ratio. We demonstrate that the enhanced PINNs are capable of identifying material parameters from both experimental one-dimensional data and synthetic full-field displacement data in a realistic regime. Since displacement data measured by, e.g., a digital image correlation (DIC) system is noisy, we additionally investigate the robustness of the method to different levels of noise.

Accurate activity location prediction is a crucial component of many mobility applications and is particularly required to develop personalized, sustainable transportation systems. Despite the widespread adoption of deep learning models, next location prediction models lack a comprehensive discussion and integration of mobility-related spatio-temporal contexts. Here, we utilize a multi-head self-attentional (MHSA) neural network that learns location transition patterns from historical location visits, their visit time and activity duration, as well as their surrounding land use functions, to infer an individual's next location. Specifically, we adopt point-of-interest data and latent Dirichlet allocation for representing locations' land use contexts at multiple spatial scales, generate embedding vectors of the spatio-temporal features, and learn to predict the next location with an MHSA network. Through experiments on two large-scale GNSS tracking datasets, we demonstrate that the proposed model outperforms other state-of-the-art prediction models, and reveal the contribution of various spatio-temporal contexts to the model's performance. Moreover, we find that the model trained on population data achieves higher prediction performance with fewer parameters than individual-level models due to learning from collective movement patterns. We also reveal mobility conducted in the recent past and one week before has the largest influence on the current prediction, showing that learning from a subset of the historical mobility is sufficient to obtain an accurate location prediction result. We believe that the proposed model is vital for context-aware mobility prediction. The gained insights will help to understand location prediction models and promote their implementation for mobility applications.

We address 2D floorplan reconstruction from 3D scans. Existing approaches typically employ heuristically designed multi-stage pipelines. Instead, we formulate floorplan reconstruction as a single-stage structured prediction task: find a variable-size set of polygons, which in turn are variable-length sequences of ordered vertices. To solve it we develop a novel Transformer architecture that generates polygons of multiple rooms in parallel, in a holistic manner without hand-crafted intermediate stages. The model features two-level queries for polygons and corners, and includes polygon matching to make the network end-to-end trainable. Our method achieves a new state-of-the-art for two challenging datasets, Structured3D and SceneCAD, along with significantly faster inference than previous methods. Moreover, it can readily be extended to predict additional information, i.e., semantic room types and architectural elements like doors and windows. Our code and models will be available at: https://github.com/ywyue/RoomFormer.

The well-documented presence of texture bias in modern convolutional neural networks has led to a plethora of algorithms that promote an emphasis on shape cues, often to support generalization to new domains. Yet, common datasets, benchmarks and general model selection strategies are missing, and there is no agreed, rigorous evaluation protocol. In this paper, we investigate difficulties and limitations when training networks with reduced texture bias. In particular, we also show that proper evaluation and meaningful comparisons between methods are not trivial. We introduce BiasBed, a testbed for texture- and style-biased training, including multiple datasets and a range of existing algorithms. It comes with an extensive evaluation protocol that includes rigorous hypothesis testing to gauge the significance of the results, despite the considerable training instability of some style bias methods. Our extensive experiments, shed new light on the need for careful, statistically founded evaluation protocols for style bias (and beyond). E.g., we find that some algorithms proposed in the literature do not significantly mitigate the impact of style bias at all. With the release of BiasBed, we hope to foster a common understanding of consistent and meaningful comparisons, and consequently faster progress towards learning methods free of texture bias. Code is available at https://github.com/D1noFuzi/BiasBed

Tasks critical to enterprise profitability, such as customer churn prediction, fraudulent account detection or customer lifetime value estimation, are often tackled by models trained on features engineered from customer data in tabular format. Application-specific feature engineering adds development, operationalization and maintenance costs over time. Recent advances in representation learning present an opportunity to simplify and generalize feature engineering across applications. When applying these advancements to tabular data researchers deal with data heterogeneity, variations in customer engagement history or the sheer volume of enterprise datasets. In this paper, we propose a novel approach to encode tabular data containing customer transactions, purchase history and other interactions into a generic representation of a customer's association with the business. We then evaluate these embeddings as features to train multiple models spanning a variety of applications. CASPR, Customer Activity Sequence-based Prediction and Representation, applies Transformer architecture to encode activity sequences to improve model performance and avoid bespoke feature engineering across applications. Our experiments at scale validate CASPR for both small and large enterprise applications.

Fine-grained population maps are needed in several domains, like urban planning, environmental monitoring, public health, and humanitarian operations. Unfortunately, in many countries only aggregate census counts over large spatial units are collected, moreover, these are not always up-to-date. We present POMELO, a deep learning model that employs coarse census counts and open geodata to estimate fine-grained population maps with 100m ground sampling distance. Moreover, the model can also estimate population numbers when no census counts at all are available, by generalizing across countries. In a series of experiments for several countries in sub-Saharan Africa, the maps produced with POMELOare in good agreement with the most detailed available reference counts: disaggregation of coarse census counts reaches R2 values of 85-89%; unconstrained prediction in the absence of any counts reaches 48-69%.

我们提出了GRASP提案网络（GP-NET），这是一种卷积神经网络模型，可以为移动操纵器生成6-DOF GRASP。为了训练GP-NET，我们合成生成一个包含深度图像和地面真相掌握信息的数据集，以供超过1400个对象。在现实世界实验中，我们使用egad！掌握基准测试，以评估两种常用算法的GP-NET，即体积抓地力网络（VGN）和在PAL TIAGO移动操纵器上进行的GRASP抓取网络（VGN）和GRASP姿势检测包（GPD）。GP-NET的掌握率为82.2％，而VGN为57.8％，GPD的成功率为63.3％。与机器人握把中最新的方法相反，GP-NET可以在不限制工作空间的情况下使用移动操纵器抓住对象，用于抓住对象，需要桌子进行分割或需要高端GPU。为了鼓励使用GP-NET，我们在https://aucoroboticsmu.github.io/gp-net/上提供ROS包以及我们的代码和预培训模型。