最近的各向同性网络，例如Convmixer和Vision Transformers，在视觉识别任务中发现了巨大的成功，匹配或胜过非方向性卷积神经网络（CNNS）。各向同性架构特别适合跨层重量共享，这是一种有效的神经网络压缩技术。在本文中，我们对各向同性网络中共享参数的方法（SPIN）进行了经验评估。我们提出了一个框架，以形式化重量分享设计决策并对此设计空间进行全面的经验评估。在我们的实验结果的指导下，我们提出了一种重量共享策略，以与仅传统缩放方法相比，在拖放和参数与准确性方面，产生一个具有更好总体效率的模型家族，例如，将Convmixer压缩为1.9倍，同时提高准确性的准确性成像网。最后，我们进行定性研究，以进一步了解各向同性体系结构中的重量共享的行为。该代码可在https://github.com/apple/ml-pin上找到。

用于移动设备的有效神经网络骨干通常针对诸如FLOPS或参数计数之类的指标进行优化。但是，这些指标在移动设备上部署时可能与网络的延迟不太相关。因此，我们通过在移动设备上部署多个移动友好网络来对不同指标进行广泛的分析。我们在最近有效的神经网络中识别和分析建筑和优化瓶颈，并提供减轻这些瓶颈的方法。为此，我们设计了一个高效的骨干莫比尼蛋白，在iPhone12上的推理时间低于1毫秒，ImageNet上的Top-1精度为75.9％。我们表明，Mobileone在高效体系结构中实现了最先进的性能，同时在移动设备上的速度更快。我们的最佳模型在38倍的速度中，在Imagenet上的性能与移动形式相似。与在类似延迟时，我们的模型在ImageNet上获得了2.3％的TOP-1精度。此外，我们表明我们的模型概括为多个任务 - 图像分类，对象检测和语义分割，与在移动设备上部署时现有的有效体系结构相比，延迟和准确性的显着提高。

对人类的逼真渲染和安息对于实现增强现实体验至关重要。我们提出了一个新颖的框架，以重建人类和场景，可以用新颖的人类姿势和景色从一个单一的野外视频中呈现。给定一个由移动摄像机捕获的视频，我们训练了两个NERF模型：人类NERF模型和一个场景NERF模型。为了训练这些模型，我们依靠现有方法来估计人类和场景的粗糙几何形状。这些粗糙的几何估计值使我们能够创建一个从观察空间到独立姿势独立的空间的翘曲场10秒的视频剪辑，并以新颖的观点以及背景提供新颖的姿势，提供人类的高质量效果。

Learned 3D representations of human faces are useful for computer vision problems such as 3D face tracking and reconstruction from images, as well as graphics applications such as character generation and animation. Traditional models learn a latent representation of a face using linear subspaces or higher-order tensor generalizations. Due to this linearity, they can not capture extreme deformations and nonlinear expressions. To address this, we introduce a versatile model that learns a non-linear representation of a face using spectral convolutions on a mesh surface. We introduce mesh sampling operations that enable a hierarchical mesh representation that captures non-linear variations in shape and expression at multiple scales within the model. In a variational setting, our model samples diverse realistic 3D faces from a multivariate Gaussian distribution. Our training data consists of 20,466 meshes of extreme expressions captured over 12 different subjects. Despite limited training data, our trained model outperforms state-of-the-art face models with 50% lower reconstruction error, while using 75% fewer parameters. We show that, replacing the expression space of an existing state-of-theart face model with our model, achieves a lower reconstruction error. Our data, model and code are available at http://coma.is.tue.mpg.de/.

We address the unsupervised learning of several interconnected problems in low-level vision: single view depth prediction, camera motion estimation, optical flow, and segmentation of a video into the static scene and moving regions. Our key insight is that these four fundamental vision problems are coupled through geometric constraints. Consequently, learning to solve them together simplifies the problem because the solutions can reinforce each other. We go beyond previous work by exploiting geometry more explicitly and segmenting the scene into static and moving regions. To that end, we introduce Competitive Collaboration, a framework that facilitates the coordinated training of multiple specialized neural networks to solve complex problems. Competitive Collaboration works much like expectation-maximization, but with neural networks that act as both competitors to explain pixels that correspond to static or moving regions, and as collaborators through a moderator that assigns pixels to be either static or independently moving. Our novel method integrates all these problems in a common framework and simultaneously reasons about the segmentation of the scene into moving objects and the static background, the camera motion, depth of the static scene structure, and the optical flow of moving objects. Our model is trained without any supervision and achieves state-of-the-art performance among joint unsupervised methods on all sub-problems. .

We learn to compute optical flow by combining a classical spatial-pyramid formulation with deep learning. This estimates large motions in a coarse-to-fine approach by warping one image of a pair at each pyramid level by the current flow estimate and computing an update to the flow. Instead of the standard minimization of an objective function at each pyramid level, we train one deep network per level to compute the flow update. Unlike the recent FlowNet approach, the networks do not need to deal with large motions; these are dealt with by the pyramid. This has several advantages. First, our Spatial Pyramid Network (SPyNet) is much simpler and 96% smaller than FlowNet in terms of model parameters. This makes it more efficient and appropriate for embedded applications. Second, since the flow at each pyramid level is small (< 1 pixel), a convolutional approach applied to pairs of warped images is appropriate. Third, unlike FlowNet, the learned convolution filters appear similar to classical spatio-temporal filters, giving insight into the method and how to improve it. Our results are more accurate than FlowNet on most standard benchmarks, suggesting a new direction of combining classical flow methods with deep learning.1 This, of course, has well-known limitations, which we discuss later.

In this paper, we propose SceNDD: a scenario-based naturalistic driving dataset that is built upon data collected from an instrumented vehicle in downtown Indianapolis. The data collection was completed in 68 driving sessions with different drivers, where each session lasted about 20--40 minutes. The main goal of creating this dataset is to provide the research community with real driving scenarios that have diverse trajectories and driving behaviors. The dataset contains ego-vehicle's waypoints, velocity, yaw angle, as well as non-ego actor's waypoints, velocity, yaw angle, entry-time, and exit-time. Certain flexibility is provided to users so that actors, sensors, lanes, roads, and obstacles can be added to the existing scenarios. We used a Joint Probabilistic Data Association (JPDA) tracker to detect non-ego vehicles on the road. We present some preliminary results of the proposed dataset and a few applications associated with it. The complete dataset is expected to be released by early 2023.

$ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while still maintaining mobility and latency. We introduce a fast and efficient convolutional neural network, ASBU-Net, for semantic segmentation of high resolution images that addresses these problems and uses no novelty layers for ease of quantization and embedded hardware support. ASBU-Net is based on a new feature extraction module, atrous space bender layer (ASBL), which is efficient in terms of computation and memory. The ASB layers form a building block that is used to make ASBNet. Since this network does not use any special layers it can be easily implemented, quantized and deployed on FPGAs and other hardware with limited memory. We present experiments on resource and accuracy trade-offs and show strong performance compared to other popular models.

For best performance, today's semantic segmentation methods use large and carefully labeled datasets, requiring expensive annotation budgets. In this work, we show that coarse annotation is a low-cost but highly effective alternative for training semantic segmentation models. Considering the urban scene segmentation scenario, we leverage cheap coarse annotations for real-world captured data, as well as synthetic data to train our model and show competitive performance compared with finely annotated real-world data. Specifically, we propose a coarse-to-fine self-training framework that generates pseudo labels for unlabeled regions of the coarsely annotated data, using synthetic data to improve predictions around the boundaries between semantic classes, and using cross-domain data augmentation to increase diversity. Our extensive experimental results on Cityscapes and BDD100k datasets demonstrate that our method achieves a significantly better performance vs annotation cost tradeoff, yielding a comparable performance to fully annotated data with only a small fraction of the annotation budget. Also, when used as pretraining, our framework performs better compared to the standard fully supervised setting.

Can we leverage the audiovisual information already present in video to improve self-supervised representation learning? To answer this question, we study various pretraining architectures and objectives within the masked autoencoding framework, motivated by the success of similar methods in natural language and image understanding. We show that we can achieve significant improvements on audiovisual downstream classification tasks, surpassing the state-of-the-art on VGGSound and AudioSet. Furthermore, we can leverage our audiovisual pretraining scheme for multiple unimodal downstream tasks using a single audiovisual pretrained model. We additionally demonstrate the transferability of our representations, achieving state-of-the-art audiovisual results on Epic Kitchens without pretraining specifically for this dataset.