本文介绍了一个新型的预训练的空间时间多对一（p-STMO）模型，用于2D到3D人类姿势估计任务。为了减少捕获空间和时间信息的困难，我们将此任务分为两个阶段：预训练（I期）和微调（II阶段）。在第一阶段，提出了一个自我监督的预训练子任务，称为蒙面姿势建模。输入序列中的人关节在空间和时间域中随机掩盖。利用denoising自动编码器的一般形式以恢复原始的2D姿势，并且编码器能够以这种方式捕获空间和时间依赖性。在第二阶段，将预训练的编码器加载到STMO模型并进行微调。编码器之后是一个多对一的框架聚合器，以预测当前帧中的3D姿势。尤其是，MLP块被用作STMO中的空间特征提取器，其性能比其他方法更好。此外，提出了一种时间下采样策略，以减少数据冗余。在两个基准上进行的广泛实验表明，我们的方法优于较少参数和较少计算开销的最先进方法。例如，我们的P-STMO模型在使用CPN作为输入的2D姿势时，在Human3.6M数据集上达到42.1mm MPJPE。同时，它为最新方法带来了1.5-7.1倍的速度。代码可在https://github.com/patrick-swk/p-stmo上找到。

Human parsing aims to partition humans in image or video into multiple pixel-level semantic parts. In the last decade, it has gained significantly increased interest in the computer vision community and has been utilized in a broad range of practical applications, from security monitoring, to social media, to visual special effects, just to name a few. Although deep learning-based human parsing solutions have made remarkable achievements, many important concepts, existing challenges, and potential research directions are still confusing. In this survey, we comprehensively review three core sub-tasks: single human parsing, multiple human parsing, and video human parsing, by introducing their respective task settings, background concepts, relevant problems and applications, representative literature, and datasets. We also present quantitative performance comparisons of the reviewed methods on benchmark datasets. Additionally, to promote sustainable development of the community, we put forward a transformer-based human parsing framework, providing a high-performance baseline for follow-up research through universal, concise, and extensible solutions. Finally, we point out a set of under-investigated open issues in this field and suggest new directions for future study. We also provide a regularly updated project page, to continuously track recent developments in this fast-advancing field: https://github.com/soeaver/awesome-human-parsing.

Recent CLIP-guided 3D optimization methods, e.g., DreamFields and PureCLIPNeRF achieve great success in zero-shot text-guided 3D synthesis. However, due to the scratch training and random initialization without any prior knowledge, these methods usually fail to generate accurate and faithful 3D structures that conform to the corresponding text. In this paper, we make the first attempt to introduce the explicit 3D shape prior to CLIP-guided 3D optimization methods. Specifically, we first generate a high-quality 3D shape from input texts in the text-to-shape stage as the 3D shape prior. We then utilize it as the initialization of a neural radiance field and then optimize it with the full prompt. For the text-to-shape generation, we present a simple yet effective approach that directly bridges the text and image modalities with a powerful text-to-image diffusion model. To narrow the style domain gap between images synthesized by the text-to-image model and shape renderings used to train the image-to-shape generator, we further propose to jointly optimize a learnable text prompt and fine-tune the text-to-image diffusion model for rendering-style image generation. Our method, namely, Dream3D, is capable of generating imaginative 3D content with better visual quality and shape accuracy than state-of-the-art methods.

To reproduce the success of text-to-image (T2I) generation, recent works in text-to-video (T2V) generation employ large-scale text-video dataset for fine-tuning. However, such paradigm is computationally expensive. Humans have the amazing ability to learn new visual concepts from just one single exemplar. We hereby study a new T2V generation problem$\unicode{x2014}$One-Shot Video Generation, where only a single text-video pair is presented for training an open-domain T2V generator. Intuitively, we propose to adapt the T2I diffusion model pretrained on massive image data for T2V generation. We make two key observations: 1) T2I models are able to generate images that align well with the verb terms; 2) extending T2I models to generate multiple images concurrently exhibits surprisingly good content consistency. To further learn continuous motion, we propose Tune-A-Video with a tailored Sparse-Causal Attention, which generates videos from text prompts via an efficient one-shot tuning of pretrained T2I diffusion models. Tune-A-Video is capable of producing temporally-coherent videos over various applications such as change of subject or background, attribute editing, style transfer, demonstrating the versatility and effectiveness of our method.

Quadrotors with the ability to perch on moving inclined surfaces can save energy and extend their travel distance by leveraging ground vehicles. Achieving dynamic perching places high demands on the performance of trajectory planning and terminal state accuracy in SE(3). However, in the perching process, uncertainties in target surface prediction, tracking control and external disturbances may cause trajectory planning failure or lead to unacceptable terminal errors. To address these challenges, we first propose a trajectory planner that considers adaptation to uncertainties in target prediction and tracking control. To facilitate this work, the reachable set of quadrotors' states is first analyzed. The states whose reachable sets possess the largest coverage probability for uncertainty targets, are defined as optimal waypoints. Subsequently, an approach to seek local optimal waypoints for static and moving uncertainty targets is proposed. A real-time trajectory planner based on optimized waypoints is developed accordingly. Secondly, thrust regulation is also implemented in the terminal attitude tracking stage to handle external disturbances. When a quadrotor's attitude is commanded to align with target surfaces, the thrust is optimized to minimize terminal errors. This makes the terminal position and velocity be controlled in closed-loop manner. Therefore, the resistance to disturbances and terminal accuracy is improved. Extensive simulation experiments demonstrate that our methods can improve the accuracy of terminal states under uncertainties. The success rate is approximately increased by $50\%$ compared to the two-end planner without thrust regulation. Perching on the rear window of a car is also achieved using our proposed heterogeneous cooperation system outdoors. This validates the feasibility and practicality of our methods.

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

The recurrent structure is a prevalent framework for the task of video super-resolution, which models the temporal dependency between frames via hidden states. When applied to real-world scenarios with unknown and complex degradations, hidden states tend to contain unpleasant artifacts and propagate them to restored frames. In this circumstance, our analyses show that such artifacts can be largely alleviated when the hidden state is replaced with a cleaner counterpart. Based on the observations, we propose a Hidden State Attention (HSA) module to mitigate artifacts in real-world video super-resolution. Specifically, we first adopt various cheap filters to produce a hidden state pool. For example, Gaussian blur filters are for smoothing artifacts while sharpening filters are for enhancing details. To aggregate a new hidden state that contains fewer artifacts from the hidden state pool, we devise a Selective Cross Attention (SCA) module, in which the attention between input features and each hidden state is calculated. Equipped with HSA, our proposed method, namely FastRealVSR, is able to achieve 2x speedup while obtaining better performance than Real-BasicVSR. Codes will be available at https://github.com/TencentARC/FastRealVSR

In contrastive self-supervised learning, the common way to learn discriminative representation is to pull different augmented "views" of the same image closer while pushing all other images further apart, which has been proven to be effective. However, it is unavoidable to construct undesirable views containing different semantic concepts during the augmentation procedure. It would damage the semantic consistency of representation to pull these augmentations closer in the feature space indiscriminately. In this study, we introduce feature-level augmentation and propose a novel semantics-consistent feature search (SCFS) method to mitigate this negative effect. The main idea of SCFS is to adaptively search semantics-consistent features to enhance the contrast between semantics-consistent regions in different augmentations. Thus, the trained model can learn to focus on meaningful object regions, improving the semantic representation ability. Extensive experiments conducted on different datasets and tasks demonstrate that SCFS effectively improves the performance of self-supervised learning and achieves state-of-the-art performance on different downstream tasks.

Recently, the success of pre-training in text domain has been fully extended to vision, audio, and cross-modal scenarios. The proposed pre-training models of different modalities are showing a rising trend of homogeneity in their model structures, which brings the opportunity to implement different pre-training models within a uniform framework. In this paper, we present TencentPretrain, a toolkit supporting pre-training models of different modalities. The core feature of TencentPretrain is the modular design. The toolkit uniformly divides pre-training models into 5 components: embedding, encoder, target embedding, decoder, and target. As almost all of common modules are provided in each component, users can choose the desired modules from different components to build a complete pre-training model. The modular design enables users to efficiently reproduce existing pre-training models or build brand-new one. We test the toolkit on text, vision, and audio benchmarks and show that it can match the performance of the original implementations.

Neural networks, especially the recent proposed neural operator models, are increasingly being used to find the solution operator of differential equations. Compared to traditional numerical solvers, they are much faster and more efficient in practical applications. However, one critical issue is that training neural operator models require large amount of ground truth data, which usually comes from the slow numerical solvers. In this paper, we propose a physics-guided data augmentation (PGDA) method to improve the accuracy and generalization of neural operator models. Training data is augmented naturally through the physical properties of differential equations such as linearity and translation. We demonstrate the advantage of PGDA on a variety of linear differential equations, showing that PGDA can improve the sample complexity and is robust to distributional shift.