Quantifying motion in 3D is important for studying the behavior of humans and other animals, but manual pose annotations are expensive and time-consuming to obtain. Self-supervised keypoint discovery is a promising strategy for estimating 3D poses without annotations. However, current keypoint discovery approaches commonly process single 2D views and do not operate in the 3D space. We propose a new method to perform self-supervised keypoint discovery in 3D from multi-view videos of behaving agents, without any keypoint or bounding box supervision in 2D or 3D. Our method uses an encoder-decoder architecture with a 3D volumetric heatmap, trained to reconstruct spatiotemporal differences across multiple views, in addition to joint length constraints on a learned 3D skeleton of the subject. In this way, we discover keypoints without requiring manual supervision in videos of humans and rats, demonstrating the potential of 3D keypoint discovery for studying behavior.

从单个图像中识别3D中的场景和对象是计算机视觉的长期目标，该目标具有机器人技术和AR/VR的应用。对于2D识别，大型数据集和可扩展解决方案已导致前所未有的进步。在3D中，现有的基准尺寸很小，并且方法专门研究几个对象类别和特定域，例如城市驾驶场景。在2D识别的成功中，我们通过引入一个称为Omni3d的大型基准来重新审视3D对象检测的任务。 OMNI3D重新排列并结合了现有的数据集，导致234K图像与超过300万个实例和97个类别相结合。由于相机内在的差异以及场景和对象类型的丰富多样性，因此3d检测到了这种规模的检测具有挑战性。我们提出了一个称为Cube R-CNN的模型，旨在以统一的方法跨相机和场景类型概括。我们表明，Cube R-CNN在较大的Omni3D和现有基准测试方面都优于先前的作品。最后，我们证明OMNI3D是一个用于3D对象识别的功能强大的数据集，表明它可以改善单数据库性能，并可以通过预训练在新的较小数据集上加速学习。

一个3D场景由一组对象组成，每个对象都有一个形状和一个布局，使其在太空中的位置。从2D图像中了解3D场景是一个重要的目标，并具有机器人技术和图形的应用。尽管最近在预测单个图像的3D形状和布局方面取得了进步，但大多数方法都依赖于3D地面真相来进行训练，这很昂贵。我们克服了这些局限性，并提出了一种方法，该方法学会预测对象的3D形状和布局，而无需任何地面真相形状或布局信息：相反，我们依靠具有2D监督的多视图图像，可以更轻松地按大规模收集。通过在3D仓库，Hypersim和扫描仪上进行的广泛实验，我们证明了我们的进近量表与逼真的图像的大型数据集相比，并与依赖3D地面真理的方法进行了比较。在Hypersim和Scannet上，如果没有可靠的3D地面真相，我们的方法优于在较小和较少的数据集上训练的监督方法。

人类可以从少量的2D视图中从3D中感知场景。对于AI代理商，只有几个图像的任何视点识别场景的能力使它们能够有效地与场景及其对象交互。在这项工作中，我们试图通过这种能力赋予机器。我们提出了一种模型，它通过将新场景的几个RGB图像进行输入，并通过将其分割为语义类别来识别新的视点中的场景。所有这一切都没有访问这些视图的RGB图像。我们将2D场景识别与隐式3D表示，并从数百个场景的多视图2D注释中学习，而无需超出相机姿势的3D监督。我们试验具有挑战性的数据集，并展示我们模型的能力，共同捕捉新颖场景的语义和几何形状，具有不同的布局，物体类型和形状。

我们提出了可区分的立体声，这是一种多视图立体方法，可从几乎没有输入视图和嘈杂摄像机中重建形状和纹理。我们将传统的立体定向和现代可区分渲染配对，以构建端到端模型，该模型可以预测具有不同拓扑和形状的物体的纹理3D网眼。我们将立体定向作为优化问题，并通过简单的梯度下降同时更新形状和相机。我们进行了广泛的定量分析，并与传统的多视图立体声技术和基于最先进的学习方法进行比较。我们展示了令人信服的重建，这些重建是在挑战现实世界的场景上，以及具有复杂形状，拓扑和纹理的大量对象类型。项目网页：https：//shubham-goel.github.io/ds/

Rapid advances in 2D perception have led to systems that accurately detect objects in real-world images. However, these systems make predictions in 2D, ignoring the 3D structure of the world. Concurrently, advances in 3D shape prediction have mostly focused on synthetic benchmarks and isolated objects. We unify advances in these two areas. We propose a system that detects objects in real-world images and produces a triangle mesh giving the full 3D shape of each detected object. Our system, called Mesh R-CNN, augments Mask R-CNN with a mesh prediction branch that outputs meshes with varying topological structure by first predicting coarse voxel representations which are converted to meshes and refined with a graph convolution network operating over the mesh's vertices and edges. We validate our mesh prediction branch on ShapeNet, where we outperform prior work on single-image shape prediction. We then deploy our full Mesh R-CNN system on Pix3D, where we jointly detect objects and predict their 3D shapes. Project page: https://gkioxari.github.io/meshrcnn/.

We present a new AI task -Embodied Question Answering (EmbodiedQA) -where an agent is spawned at a random location in a 3D environment and asked a question ('What color is the car?'). In order to answer, the agent must first intelligently navigate to explore the environment, gather information through first-person (egocentric) vision, and then answer the question ('orange'). This challenging task requires a range of AI skills -active perception, language understanding, goal-driven navigation, commonsense reasoning, and grounding of language into actions. In this work, we develop the environments, end-to-end-trained reinforcement learning agents, and evaluation protocols for EmbodiedQA.

Merging satellite products and ground-based measurements is often required for obtaining precipitation datasets that simultaneously cover large regions with high density and are more accurate than pure satellite precipitation products. Machine and statistical learning regression algorithms are regularly utilized in this endeavour. At the same time, tree-based ensemble algorithms for regression are adopted in various fields for solving algorithmic problems with high accuracy and low computational cost. The latter can constitute a crucial factor for selecting algorithms for satellite precipitation product correction at the daily and finer time scales, where the size of the datasets is particularly large. Still, information on which tree-based ensemble algorithm to select in such a case for the contiguous United States (US) is missing from the literature. In this work, we conduct an extensive comparison between three tree-based ensemble algorithms, specifically random forests, gradient boosting machines (gbm) and extreme gradient boosting (XGBoost), in the context of interest. We use daily data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and the IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets. We also use earth-observed precipitation data from the Global Historical Climatology Network daily (GHCNd) database. The experiments refer to the entire contiguous US and additionally include the application of the linear regression algorithm for benchmarking purposes. The results suggest that XGBoost is the best-performing tree-based ensemble algorithm among those compared. They also suggest that IMERG is more useful than PERSIANN in the context investigated.

Many, if not most, systems of interest in science are naturally described as nonlinear dynamical systems (DS). Empirically, we commonly access these systems through time series measurements, where often we have time series from different types of data modalities simultaneously. For instance, we may have event counts in addition to some continuous signal. While by now there are many powerful machine learning (ML) tools for integrating different data modalities into predictive models, this has rarely been approached so far from the perspective of uncovering the underlying, data-generating DS (aka DS reconstruction). Recently, sparse teacher forcing (TF) has been suggested as an efficient control-theoretic method for dealing with exploding loss gradients when training ML models on chaotic DS. Here we incorporate this idea into a novel recurrent neural network (RNN) training framework for DS reconstruction based on multimodal variational autoencoders (MVAE). The forcing signal for the RNN is generated by the MVAE which integrates different types of simultaneously given time series data into a joint latent code optimal for DS reconstruction. We show that this training method achieves significantly better reconstructions on multimodal datasets generated from chaotic DS benchmarks than various alternative methods.

We present hierarchical policy blending as optimal transport (HiPBOT). This hierarchical framework adapts the weights of low-level reactive expert policies, adding a look-ahead planning layer on the parameter space of a product of expert policies and agents. Our high-level planner realizes a policy blending via unbalanced optimal transport, consolidating the scaling of underlying Riemannian motion policies, effectively adjusting their Riemannian matrix, and deciding over the priorities between experts and agents, guaranteeing safety and task success. Our experimental results in a range of application scenarios from low-dimensional navigation to high-dimensional whole-body control showcase the efficacy and efficiency of HiPBOT, which outperforms state-of-the-art baselines that either perform probabilistic inference or define a tree structure of experts, paving the way for new applications of optimal transport to robot control. More material at https://sites.google.com/view/hipobot