The use of larger antenna arrays at higher frequency bands is envisioned in the beyond 5G networks. This takes advantage of the near-field propagation regime where the wavefront is no longer plane but spherical, bringing both new opportunities and challenges for the high-precision positioning. In this paper, a generic near-field positioning model with different observation capabilities for three electric fields (vector, scalar and overall scalar electric field) is proposed. For these three electric field types, the Cram\'er-Rao bound (CRB) is adopted to evaluate the achievable estimation accuracy. The expressions of the CRBs using different electric field observations are derived by combining electromagnetic theory with estimation theory. Closed-form expressions can be further obtained if the terminal is located on the central perpendicular line (CPL) of the receiving antenna surface. In addition, the above discussions are extended to the system with multiple distributed receiving antennas under the CPL assumption. The CRBs using various electric fields in this case are derived and the effect of different numbers of receiving antennas on estimation accuracy is investigated. Numerical results are provided to quantify the CRBs and validate the analytical results. Also, the impact of various system parameters, including different electric fields and multiple antennas, on the near-field positioning performance is evaluated.
Existing learning-based frame interpolation algorithms extract consecutive frames from high-speed natural videos to train the model. Compared to natural videos, cartoon videos are usually in a low frame rate. Besides, the motion between consecutive cartoon frames is typically nonlinear, which breaks the linear motion assumption of interpolation algorithms. Thus, it is unsuitable for generating a training set directly from cartoon videos. For better adapting frame interpolation algorithms from nature video to animation video, we present AutoFI, a simple and effective method to automatically render training data for deep animation video interpolation. AutoFI takes a layered architecture to render synthetic data, which ensures the assumption of linear motion. Experimental results show that AutoFI performs favorably in training both DAIN and ANIN. However, most frame interpolation algorithms will still fail in error-prone areas, such as fast motion or large occlusion. Besides AutoFI, we also propose a plug-and-play sketch-based post-processing module, named SktFI, to refine the final results using user-provided sketches manually. With AutoFI and SktFI, the interpolated animation frames show high perceptual quality.
Equipped with a wide span of sensors, predominant autonomous driving solutions are becoming more modular-oriented for safe system design. Though these sensors have laid a solid foundation, most massive-production solutions up to date still fall into L2 phase. Among these, Comma.ai comes to our sight, claiming one $999 aftermarket device mounted with a single camera and board inside owns the ability to handle L2 scenarios. Together with open-sourced software of the entire system released by Comma.ai, the project is named Openpilot. Is it possible? If so, how is it made possible? With curiosity in mind, we deep-dive into Openpilot and conclude that its key to success is the end-to-end system design instead of a conventional modular framework. The model is briefed as Supercombo, and it can predict the ego vehicle's future trajectory and other road semantics on the fly from monocular input. Unfortunately, the training process and massive amount of data to make all these work are not publicly available. To achieve an intensive investigation, we try to reimplement the training details and test the pipeline on public benchmarks. The refactored network proposed in this work is referred to as OP-Deepdive. For a fair comparison of our version to the original Supercombo, we introduce a dual-model deployment scheme to test the driving performance in the real world. Experimental results on nuScenes, Comma2k19, CARLA, and in-house realistic scenarios verify that a low-cost device can indeed achieve most L2 functionalities and be on par with the original Supercombo model. In this report, we would like to share our latest findings, shed some light on the new perspective of end-to-end autonomous driving from an industrial product-level side, and potentially inspire the community to continue improving the performance. Our code, benchmarks are at https://github.com/OpenPerceptionX/Openpilot-Deepdive.
Current end-to-end autonomous driving methods either run a controller based on a planned trajectory or perform control prediction directly, which have spanned two separately studied lines of research. Seeing their potential mutual benefits to each other, this paper takes the initiative to explore the combination of these two well-developed worlds. Specifically, our integrated approach has two branches for trajectory planning and direct control, respectively. The trajectory branch predicts the future trajectory, while the control branch involves a novel multi-step prediction scheme such that the relationship between current actions and future states can be reasoned. The two branches are connected so that the control branch receives corresponding guidance from the trajectory branch at each time step. The outputs from two branches are then fused to achieve complementary advantages. Our results are evaluated in the closed-loop urban driving setting with challenging scenarios using the CARLA simulator. Even with a monocular camera input, the proposed approach ranks $first$ on the official CARLA Leaderboard, outperforming other complex candidates with multiple sensors or fusion mechanisms by a large margin. The source code and data will be made publicly available at https://github.com/OpenPerceptionX/TCP.
Deep neural networks are susceptible to label noise. Existing methods to improve robustness, such as meta-learning and regularization, usually require significant change to the network architecture or careful tuning of the optimization procedure. In this work, we propose a simple hierarchical approach that incorporates a label hierarchy when training the deep learning models. Our approach requires no change of the network architecture or the optimization procedure. We investigate our hierarchical network through a wide range of simulated and real datasets and various label noise types. Our hierarchical approach improves upon regular deep neural networks in learning with label noise. Combining our hierarchical approach with pre-trained models achieves state-of-the-art performance in real-world noisy datasets.
In time-division duplexing (TDD) millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) systems, the reciprocity mismatch severely degrades the performance of the hybrid beamforming (HBF). In this work, to mitigate the detrimental effect of the reciprocity mismatch, we investigate reciprocity calibration for the mmWave-HBF system with a fully-connected phase shifter network. To reduce the overhead and computational complexity of reciprocity calibration, we first decouple digital radio frequency (RF) chains and analog RF chains with beamforming design. Then, the entire calibration problem of the HBF system is equivalently decomposed into two subproblems corresponding to the digital-chain calibration and analog-chain calibration. To solve the calibration problems efficiently, a closed-form solution to the digital-chain calibration problem is derived, while an iterative-alternating optimization algorithm for the analog-chain calibration problem is proposed. To measure the performance of the proposed algorithm, we derive the Cram\'er-Rao lower bound on the errors in estimating mismatch coefficients. The results reveal that the estimation errors of mismatch coefficients of digital and analog chains are uncorrelated, and that the mismatch coefficients of receive digital chains can be estimated perfectly. Simulation results are presented to validate the analytical results and to show the performance of the proposed calibration approach.
Methods for 3D lane detection have been recently proposed to address the issue of inaccurate lane layouts in many autonomous driving scenarios (uphill/downhill, bump, etc.). Previous work struggled in complex cases due to their simple designs of the spatial transformation between front view and bird's eye view (BEV) and the lack of a realistic dataset. Towards these issues, we present PersFormer: an end-to-end monocular 3D lane detector with a novel Transformer-based spatial feature transformation module. Our model generates BEV features by attending to related front-view local regions with camera parameters as a reference. PersFormer adopts a unified 2D/3D anchor design and an auxiliary task to detect 2D/3D lanes simultaneously, enhancing the feature consistency and sharing the benefits of multi-task learning. Moreover, we release one of the first large-scale real-world 3D lane datasets, which is called OpenLane, with high-quality annotation and scenario diversity. OpenLane contains 200,000 frames, over 880,000 instance-level lanes, 14 lane categories, along with scene tags and the closed-in-path object annotations to encourage the development of lane detection and more industrial-related autonomous driving methods. We show that PersFormer significantly outperforms competitive baselines in the 3D lane detection task on our new OpenLane dataset as well as Apollo 3D Lane Synthetic dataset, and is also on par with state-of-the-art algorithms in the 2D task on OpenLane. The project page is available at https://github.com/OpenPerceptionX/PersFormer_3DLane and OpenLane dataset is provided at https://github.com/OpenPerceptionX/OpenLane.
Sparse Transformer has recently attracted a lot of attention since the ability for reducing the quadratic dependency on the sequence length. We argue that two factors, information bottleneck sensitivity and inconsistency between different attention topologies, could affect the performance of the Sparse Transformer. This paper proposes a well-designed model named ERNIE-Sparse. It consists of two distinctive parts: (i) Hierarchical Sparse Transformer (HST) to sequentially unify local and global information. (ii) Self-Attention Regularization (SAR) method, a novel regularization designed to minimize the distance for transformers with different attention topologies. To evaluate the effectiveness of ERNIE-Sparse, we perform extensive evaluations. Firstly, we perform experiments on a multi-modal long sequence modeling task benchmark, Long Range Arena (LRA). Experimental results demonstrate that ERNIE-Sparse significantly outperforms a variety of strong baseline methods including the dense attention and other efficient sparse attention methods and achieves improvements by 2.77% (57.78% vs. 55.01%). Secondly, to further show the effectiveness of our method, we pretrain ERNIE-Sparse and verified it on 3 text classification and 2 QA downstream tasks, achieve improvements on classification benchmark by 0.83% (92.46% vs. 91.63%), on QA benchmark by 3.24% (74.67% vs. 71.43%). Experimental results continue to demonstrate its superior performance.
Animated avatars, which look and talk like humans, are iconic visions of the future of AI-powered systems. Through many sci-fi movies we are acquainted with the idea of speaking to such virtual personalities as if they were humans. Today, we talk more and more to machines like Apple's Siri, e.g., to ask them for the weather forecast. However, when asked for recommendations, e.g., for a restaurant to go to, the limitations of such devices quickly become obvious. They do not engage in a conversation to find out what we might prefer, they often do not provide explanations for what they recommend, and they may have difficulties remembering what was said one minute earlier. Conversational recommender systems promise to address these limitations. In this paper, we review existing approaches to build such systems, which developments we observe today, which challenges are still open and why the development of conversational recommenders represents one of the next grand challenges of AI.
The surface defect detection method based on visual perception has been widely used in industrial quality inspection. Because defect data are not easy to obtain and the annotation of a large number of defect data will waste a lot of manpower and material resources. Therefore, this paper reviews the methods of surface defect detection of industrial products based on a small number of labeled data, and this method is divided into traditional image processing-based industrial product surface defect detection methods and deep learning-based industrial product surface defect detection methods suitable for a small number of labeled data. The traditional image processing-based industrial product surface defect detection methods are divided into statistical methods, spectral methods and model methods. Deep learning-based industrial product surface defect detection methods suitable for a small number of labeled data are divided into based on data augmentation, based on transfer learning, model-based fine-tuning, semi-supervised, weak supervised and unsupervised.