Graphic layout designs play an essential role in visual communication. Yet handcrafting layout designs are skill-demanding, time-consuming, and non-scalable to batch production. Although generative models emerge to make design automation no longer utopian, it remains non-trivial to customize designs that comply with designers' multimodal desires, i.e., constrained by background images and driven by foreground contents. In this study, we propose \textit{LayoutDETR} that inherits the high quality and realism from generative modeling, in the meanwhile reformulating content-aware requirements as a detection problem: we learn to detect in a background image the reasonable locations, scales, and spatial relations for multimodal elements in a layout. Experiments validate that our solution yields new state-of-the-art performance for layout generation on public benchmarks and on our newly-curated ads banner dataset. For practical usage, we build our solution into a graphical system that facilitates user studies. We demonstrate that our designs attract more subjective preference than baselines by significant margins. Our code, models, dataset, graphical system, and demos are available at https://github.com/salesforce/LayoutDETR.

We study a challenging task, conditional human motion generation, which produces plausible human motion sequences according to various conditional inputs, such as action classes or textual descriptors. Since human motions are highly diverse and have a property of quite different distribution from conditional modalities, such as textual descriptors in natural languages, it is hard to learn a probabilistic mapping from the desired conditional modality to the human motion sequences. Besides, the raw motion data from the motion capture system might be redundant in sequences and contain noises; directly modeling the joint distribution over the raw motion sequences and conditional modalities would need a heavy computational overhead and might result in artifacts introduced by the captured noises. To learn a better representation of the various human motion sequences, we first design a powerful Variational AutoEncoder (VAE) and arrive at a representative and low-dimensional latent code for a human motion sequence. Then, instead of using a diffusion model to establish the connections between the raw motion sequences and the conditional inputs, we perform a diffusion process on the motion latent space. Our proposed Motion Latent-based Diffusion model (MLD) could produce vivid motion sequences conforming to the given conditional inputs and substantially reduce the computational overhead in both the training and inference stages. Extensive experiments on various human motion generation tasks demonstrate that our MLD achieves significant improvements over the state-of-the-art methods among extensive human motion generation tasks, with two orders of magnitude faster than previous diffusion models on raw motion sequences.

Early-exiting dynamic neural networks (EDNN), as one type of dynamic neural networks, has been widely studied recently. A typical EDNN has multiple prediction heads at different layers of the network backbone. During inference, the model will exit at either the last prediction head or an intermediate prediction head where the prediction confidence is higher than a predefined threshold. To optimize the model, these prediction heads together with the network backbone are trained on every batch of training data. This brings a train-test mismatch problem that all the prediction heads are optimized on all types of data in training phase while the deeper heads will only see difficult inputs in testing phase. Treating training and testing inputs differently at the two phases will cause the mismatch between training and testing data distributions. To mitigate this problem, we formulate an EDNN as an additive model inspired by gradient boosting, and propose multiple training techniques to optimize the model effectively. We name our method BoostNet. Our experiments show it achieves the state-of-the-art performance on CIFAR100 and ImageNet datasets in both anytime and budgeted-batch prediction modes. Our code is released at https://github.com/SHI-Labs/Boosted-Dynamic-Networks.

Intent classification and slot filling are two core tasks in natural language understanding (NLU). The interaction nature of the two tasks makes the joint models often outperform the single designs. One of the promising solutions, called BERT (Bidirectional Encoder Representations from Transformers), achieves the joint optimization of the two tasks. BERT adopts the wordpiece to tokenize each input token into multiple sub-tokens, which causes a mismatch between the tokens and the labels lengths. Previous methods utilize the hidden states corresponding to the first sub-token as input to the classifier, which limits performance improvement since some hidden semantic informations is discarded in the fine-tune process. To address this issue, we propose a novel joint model based on BERT, which explicitly models the multiple sub-tokens features after wordpiece tokenization, thereby generating the context features that contribute to slot filling. Specifically, we encode the hidden states corresponding to multiple sub-tokens into a context vector via the attention mechanism. Then, we feed each context vector into the slot filling encoder, which preserves the integrity of the sentence. Experimental results demonstrate that our proposed model achieves significant improvement on intent classification accuracy, slot filling F1, and sentence-level semantic frame accuracy on two public benchmark datasets. The F1 score of the slot filling in particular has been improved from 96.1 to 98.2 (2.1% absolute) on the ATIS dataset.

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper.

在卷积神经网络（CNN）的动力下，医学图像分类迅速发展。由于卷积内核的接受场的固定尺寸，很难捕获医学图像的全局特征。尽管基于自发的变压器可以对远程依赖性进行建模，但它具有很高的计算复杂性，并且缺乏局部电感偏见。许多研究表明，全球和本地特征对于图像分类至关重要。但是，医学图像具有许多嘈杂，分散的特征，类内的变化和类间的相似性。本文提出了三个分支分层的多尺度特征融合网络结构，称为医学图像分类为新方法。它可以融合多尺度层次结构的变压器和CNN的优势，而不会破坏各自的建模，从而提高各种医学图像的分类精度。局部和全局特征块的平行层次结构旨在有效地提取各种语义尺度的本地特征和全局表示，并灵活地在不同的尺度上建模，并与图像大小相关的线性计算复杂性。此外，自适应分层特征融合块（HFF块）旨在全面利用在不同层次级别获得的功能。 HFF块包含空间注意力，通道注意力，残留的倒置MLP和快捷方式，以在每个分支的各个规模特征之间适应融合语义信息。我们在ISIC2018数据集上提出的模型的准确性比基线高7.6％，COVID-19数据集的准确性为21.5％，Kvasir数据集的准确性为10.4％。与其他高级模型相比，HIFUSE模型表现最好。我们的代码是开源的，可从https://github.com/huoxiangzuo/hifuse获得。

从单眼视频中进行的3D人姿势估计最近看到了显着改善。但是，大多数最先进的方法都是基于运动学的，它容易出现具有明显伪影的物理上不可信的运动。当前基于动态的方法可以预测物理上合理的运动，但仅限于具有静态相机视图的简单场景。在这项工作中，我们介绍了D＆D（从动态相机中学习人类动力学），该法律利用物理定律使用移动的摄像机从野外视频中重建3D人类运动。 D＆D引入了惯性力控制（IFC），以考虑动态摄像机的惯性力来解释非惯性局部框架中的3D人运动。为了学习有限注释的接地接触，我们开发了概率接触扭矩（PCT），该概率是通过与接触概率的可区分抽样计算的，并用于生成运动。接触状态可以通过鼓励模型产生正确的动作来弱监督。此外，我们提出了一个细心的PD控制器，该控制器使用时间信息来调整目标姿势状态，以获得平稳而准确的姿势控制。我们的方法完全是基于神经的，并且在物理引擎中没有离线优化或模拟的情况下运行。大规模3D人体运动基准的实验证明了D＆D的有效性，在该基于最新的运动学基于动力学和基于动力学的方法的情况下，我们表现出卓越的性能。代码可从https://github.com/jeffsjtu/dnd获得

尽管在各种应用中取得了突出的性能，但点云识别模型经常遭受自然腐败和对抗性扰动的困扰。在本文中，我们深入研究了点云识别模型的一般鲁棒性，并提出了点云对比对抗训练（PointCat）。 PointCat的主要直觉是鼓励目标识别模型缩小清洁点云和损坏点云之间的决策差距。具体而言，我们利用有监督的对比损失来促进识别模型提取的超晶体特征的对齐和均匀性，并设计一对带有动态原型指南的集中式损失，以避免这些特征与其属于其属于其归属类别群的偏离。为了提供更具挑战性的损坏点云，我们对噪声生成器以及从头开始的识别模型进行了对手训练，而不是将基于梯度的攻击用作内部循环，例如以前的对手训练方法。全面的实验表明，在包括各种损坏的情况下，所提出的PointCat优于基线方法，并显着提高不同点云识别模型的稳健性，包括各向同性点噪声，LIDAR模拟的噪声，随机点掉落和对抗性扰动。

本文提出了一种新颖的统一特征优化（UFO）范式，用于训练和在现实世界和大规模场景下进行深层模型，这需要集合多个AI功能。不明飞行物的目标是通过对所有任务进行大规模预修。与众所周知的基础模型相比，UFO具有两个不同的重点，即相对较小的模型大小，没有适应性成本：1）UFO以多任务学习方式将广泛的任务挤入中等尺寸的统一模型中并在转移到下游任务时进一步修剪模型大小。 2）不明飞行物不强调转移到新任务。相反，它旨在使修剪模型专门用于一个或多个已经看到的任务。有了这两个特征，UFO为灵活的部署提供了极大的便利，同时保持了大规模预处理的好处。 UFO的一个关键优点是修剪过程不仅可以减少模型的大小和推理消耗，而且还提高了某些任务的准确性。具体而言，UFO考虑了多任务培训，并对统一模型产生了两倍的影响：一些密切相关的任务具有相互利益，而某些任务相互冲突。不明飞行物设法通过新颖的网络体系结构搜索（NAS）方法来减少冲突并保留相互利益。对各种深度表示学习任务（即面部识别，人重新识别，车辆重新识别和产品检索）的实验表明，从UFO中修剪的模型比单件任务训练的对应物更高，但却具有更高的准确性较小的型号大小，验证不明飞行物的概念。此外，UFO还支持发布170亿个参数计算机视觉（CV）基础模型，该模型是该行业中最大的CV模型。

近年来，面部语义指导（包括面部地标，面部热图和面部解析图）和面部生成对抗网络（GAN）近年来已广泛用于盲面修复（BFR）。尽管现有的BFR方法在普通案例中取得了良好的性能，但这些解决方案在面对严重降解和姿势变化的图像时具有有限的弹性（例如，在现实世界情景中看起来右，左看，笑等）。在这项工作中，我们提出了一个精心设计的盲人面部修复网络，具有生成性面部先验。所提出的网络主要由非对称编解码器和stylegan2先验网络组成。在非对称编解码器中，我们采用混合的多路残留块（MMRB）来逐渐提取输入图像的弱纹理特征，从而可以更好地保留原始面部特征并避免过多的幻想。 MMRB也可以在其他网络中插入插件。此外，多亏了StyleGAN2模型的富裕和多样化的面部先验，我们采用了微调的方法来灵活地恢复自然和现实的面部细节。此外，一种新颖的自我监督训练策略是专门设计用于面部修复任务的，以使分配更接近目标并保持训练稳定性。关于合成和现实世界数据集的广泛实验表明，我们的模型在面部恢复和面部超分辨率任务方面取得了卓越的表现。